Parking location management device, parking location management method, and vehicle device

Abstract

When a plurality of vehicles are notified by a map server (2) that a parked vehicle is located near a roadside parking location, the plurality of vehicles upload data indicating the behavior of the vehicles and sensing information from surrounding monitoring sensors as a roadside parking location report to the map server (2). The map server (2) determines the appearance point of the parked vehicle based on the vehicle behavior data and sensing information provided by the plurality of vehicles, and registers the point as the roadside parking location in a roadside parking location DB (251). In addition, the map server (2) determines that the parked vehicle has disappeared (moved) based on the fact that no action such as avoiding the parked vehicle is observed in the behavior of the vehicles passing near the roadside parking location or that no parked vehicle is detected by the surrounding monitoring sensors.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application is based on patent application No. 2020-107960 filed in Japan on June 23, 2020, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0003] The present disclosure relates to a technology for detecting obstacles that may hinder the passage of vehicles, and more particularly to a parking spot management device and a parking spot management method for managing position information of spots where vehicles are parked on a road. Background Art

[0004] As a technology for detecting obstacles on the road, for example, Patent Document 1 discloses a structure for detecting animal carcasses, fallen trees, and objects dropped from a moving vehicle from images taken by a vehicle-mounted camera. As examples of dropped objects, boxes, ladders, skis, and the like are assumed. In addition, Patent Document 1 discloses a structure in which a vehicle uses a vehicle-mounted camera to confirm whether a dropped object notified from a server still remains, and sends the result back to the server, and the server updates the status of the dropped object based on the confirmation result from the vehicle. In addition, Patent Document 1 also mentions a structure for roughly predicting the time required to remove an obstacle based on the type of dropped object and distributing the predicted time.

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2019-40539

[0006] The obstacles assumed in Patent Document 1 are only objects that should be removed by road managers, such as animal carcasses, fallen trees, and objects dropped from vehicles, that is, stationary objects that do not move autonomously. In Patent Document 1, vehicles parked or stopped (hereinafter referred to as parked) near the end of a road are not assumed to be obstacles. Unlike stationary objects such as fallen objects, vehicles parked on the road are restarted by the user of the vehicle, so their existence state may change relatively dynamically compared to fallen objects. In other words, the characteristics such as the existence period of a parked vehicle are different from those of the obstacles assumed in Patent Document 1.

[0007] Furthermore, the presence of parked vehicles on the road can hinder services such as autonomous driving functions on regular roads. If the location of parked vehicles could be obtained as map data, it would be possible to develop driving plans that avoid these locations and implement preemptive handovers. In other words, to practically implement autonomous driving technology and improve safety and convenience, technology that can quickly detect locations where vehicles are parked on the road is required. Summary of the Invention

[0008] The present disclosure has been made based on this situation, and an object of the present disclosure is to provide a parking spot management device, a parking spot management method, and a vehicle device that can detect a spot where a vehicle is parked on a road.

[0009] The parking location management device for achieving this purpose comprises: a judgment material acquisition unit, which acquires at least one of vehicle behavior data representing the behavior of at least one vehicle and sensing information of a surrounding monitoring sensor mounted on the vehicle in correspondence with position information; a parking location detection unit, which detects a location on the road where a vehicle is parked based on the information acquired by the judgment material acquisition unit; and an existence state determination unit, which determines whether there is still a vehicle at the location where the vehicle is parked, i.e., the parking location, detected by the parking location detection unit, based on the information acquired by the judgment material acquisition unit.

[0010] According to the above configuration, the location where the vehicle is parked is detected based on the behavior data or sensing information of at least one vehicle passing through the location where the vehicle is parked.

[0011] In addition, the parking location management method for achieving the above-mentioned purpose is a method for managing location information of parked vehicles on the road, which is executed using at least one processor and includes: a judgment material acquisition step, which acquires at least any one of vehicle behavior data representing the behavior of at least one vehicle and sensing information of a peripheral monitoring sensor mounted on the vehicle in a corresponding relationship with the position information; a parking location detection step, which detects the location where the vehicle is parked on the road based on the information acquired in the judgment material acquisition step; and an existence determination step, which determines whether there is still a vehicle at the location where the vehicle is parked detected in the parking location detection step, that is, the parking location, based on the information acquired in the judgment material acquisition step.

[0012] According to the above method, a location where a vehicle is parked on a road can be detected based on information from vehicles traveling on the road, such as vehicle behavior data and / or sensing information from a surrounding monitoring sensor.

[0013] Furthermore, the vehicle device for achieving the above-mentioned purpose is a vehicle device for sending information about a location where a vehicle is parked on a road, i.e., a parking location, to a predetermined server, and comprises: a stopped vehicle information acquisition unit for acquiring information about other vehicles stopped on the road, i.e., stopped vehicles, based on an input signal from a surrounding monitoring sensor mounted on the vehicle; a parking determination unit for determining whether the stopped vehicle is equivalent to a roadside parked vehicle parked on the road, or equivalent to a temporarily parked vehicle that remains in a movable state, based on the information about the stopped vehicle acquired by the stopped vehicle information acquisition unit; and a report processing unit for sending a data set indicating a location where a roadside parked vehicle exists to the server when there is a stopped vehicle that the parking determination unit determines to be a roadside parked vehicle.

[0014] With the aforementioned vehicle device, information about locations where vehicles are likely to be parked on the road is collected on a server. Therefore, the server can detect parked locations based on the collected information from multiple vehicles. In other words, the location and status of parked vehicles can be identified in real time.

[0015] In addition, the reference numerals in parentheses in the claims indicate the correspondence with specific elements described in the embodiment described later as one aspect, and do not limit the technical scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a diagram for explaining the configuration of the roadside parking information distribution system 100 .

[0017] Figure 2 It is a block diagram showing the configuration of the in-vehicle system 1 .

[0018] Figure 3 It is a block diagram showing the configuration of the positioner 14 .

[0019] Figure 4 1 is a diagram showing an example of a roadside parked vehicle notification image 80 .

[0020] Figure 5 It is a block diagram showing the structure of the map cooperation device 50.

[0021] Figure 6 This is a flowchart showing an example of upload processing.

[0022] Figure 7 This is a diagram for explaining the range of vehicle behavior data included in a roadside parking location report.

[0023] Figure 8 This is a flowchart showing an example of upload processing.

[0024] Figure 9 This is a flowchart showing an example of the operation of the map cooperation device 50.

[0025] Figure 10 This is a flowchart showing an example of the operation of the map cooperation device 50.

[0026] Figure 11 It is a block diagram showing the structure of the map server 2.

[0027] Figure 12 2 is a block diagram showing the functions of the map server 2 provided by the server processor 21 .

[0028] Figure 13 This is a flowchart for explaining the processing in the map server 2.

[0029] Figure 14 This is a diagram for explaining the operation of the appearance determination unit G31.

[0030] Figure 15 This is a diagram showing an example of a criterion for the presence determination unit G31 to determine that an obstacle is present.

[0031] Figure 16 This is a diagram showing an example of a criterion for the disappearance determination unit G32 to determine that an obstacle has disappeared.

[0032] Figure 17 This is a flowchart showing an example of vehicle control using roadside parking location information.

[0033] Figure 18 It is a diagram for explaining the effects of the roadside parking information distribution system 100 .

[0034] Figure 19 This is a flowchart for explaining a process of determining whether a detected stopped vehicle corresponds to a roadside parked vehicle using map data.

[0035] Figure 20 This is a diagram for explaining a configuration in which the line of sight of a driver's seat passenger is used as a criterion for determining the presence or absence of a vehicle parked on the roadside.

[0036] Figure 21 This is a flowchart showing an example of operation when the host vehicle is parked in a gap in a parking row.

[0037] Figure 22 This is a diagram showing an example of a criterion used when the roadside parking presence / absence determination unit F51 calculates the detection reliability.

[0038] Figure 23 This is a diagram showing a modified example of the map server 2.

[0039] Figure 24This is a diagram schematically showing an example of a calculation rule for the accuracy of the actual presence of roadside parked vehicles by the map server 2 .

[0040] Figure 25 This is a diagram for explaining a configuration for grouping and distributing roadside stops.

[0041] Figure 26 This is a diagram for explaining the grouping of roadside stops.

[0042] Figure 27 This is a diagram for explaining an example of a rule for grouping roadside stops.

[0043] Figure 28 This is a diagram for explaining the operation of the grouping unit G34.

[0044] Figure 29 This is a diagram showing the configuration of the map cooperation device 50 that uploads information related to traffic congestion areas.

[0045] Figure 30 This figure shows the configuration of the map server 2 that detects and distributes the end of a traffic congestion section based on reports from vehicles.

[0046] Figure 31 This diagram shows the configuration of a system that dynamically sets and utilizes sections where automated driving is not permitted based on roadside parking location information and traffic congestion information. DETAILED DESCRIPTION

[0047] Hereinafter, embodiments of the present disclosure will be described using the drawings. Figure 1 1 is a diagram showing an example of a schematic configuration of the roadside parking information distribution system 100 of the present disclosure. Figure 1 As shown, a roadside parking information distribution system 100 includes multiple in-vehicle systems 1 and a map server 2, each installed in a plurality of vehicles Ma and Mb. The roadside parking information distribution system 100 uses the map server 2 to distribute information about locations where vehicles are parked on a typical road. Parked vehicles here refer to vehicles parked along the roadside or at the roadside edge of a typical road. The term "parked" can be replaced with "parked" or "parked." Locations with parked vehicles, i.e., roadside parking locations, correspond to parked vehicle locations.

[0048] In this disclosure, to distinguish them from parked vehicles, vehicles that have temporarily stopped to wait for a signal, etc., are referred to as temporarily parked vehicles. In situations where no distinction is made between parked vehicles and temporarily parked vehicles, they are referred to as stopped vehicles or parked vehicles. In other words, the concepts of stopped vehicles and parked vehicles encompass both parked vehicles and temporarily parked vehicles. A temporarily parked vehicle refers to a vehicle that has stopped to wait for a signal, for pedestrians / animals to cross, for oncoming vehicles / trains to pass, etc. (hereinafter referred to as "waiting for a signal, etc."). A temporarily parked vehicle is a vehicle that remains in a state of immediate mobility and can be expected to move to another location within a few minutes, or at most within five minutes. On the other hand, a parked vehicle conceptually refers to a vehicle that is expected to remain stationary at the same location for more than five minutes. For example, a parked vehicle is a vehicle with no passengers in the vehicle. A parked vehicle is equivalent to a stopped vehicle that should not be queued behind the vehicle and should be avoided and continued to move, while a temporarily parked vehicle is equivalent to a stopped vehicle that should be queued behind the vehicle. The end of a traffic jam is considered here as an example, similar to temporarily parked vehicles, but is not limited to this. If traffic congestion occurs only in a portion of a lane, such as a right-turn lane or a left-turn lane, and the vehicle does not need to travel in that lane, the vehicle at the end of the traffic jam can be included in the list of parked vehicles to be avoided. The end of a traffic jam can also be treated as a parking location.

[0049] In addition, as an example, it is assumed here that the roadside parking information distribution system 100 is used in an area with left-hand traffic, and the lane at the left end of the lanes with the same direction of travel is called the first lane. In addition, it is assumed that when a vehicle is parked on a road that is not a one-way street, it is parked at the end of the road on the left side of the direction of travel. When used in an area with right-hand traffic, the structure of the present disclosure can be implemented by reversing the above-mentioned left and right. For example, in an area with right-hand traffic, the first lane refers to the lane at the right end of the lanes with the same direction of travel. The roadside parking information distribution system 100 described below can be changed to suit the traffic regulations of the area where it is used. In addition, the parking position on a one-way road is not limited to either the left or right side of the road.

[0050] In addition, Figure 1 For convenience, only two vehicles, Ma and Mb, are shown as vehicles equipped with the in-vehicle system 1, but in reality, there may be three or more. The in-vehicle system 1 can be installed on any road-capable vehicle. Vehicles Ma and Mb may be two-wheeled vehicles, three-wheeled vehicles, or other vehicles in addition to four-wheeled vehicles. Bicycles with prime movers can also be included in two-wheeled vehicles. From now on, with respect to the in-vehicle system 1, the vehicle equipped with the system (i.e., the vehicle itself) will also be referred to as the "host vehicle."

[0051] <Overview of overall structure>

[0052] The vehicle-mounted system 1 mounted on each vehicle is configured to be wirelessly connectable to a wide area communication network 3. The wide area communication network 3 here refers to a public communication network provided by a telecommunications operator, such as a mobile phone network or the Internet. Figure 1 The base station 4 shown is a wireless base station for connecting the in-vehicle system 1 to the wide area communication network 3 .

[0053] Each vehicle-mounted system 1 sends a communication data packet indicating the status of the vehicle, i.e., a vehicle status report, to the map server 2 via the base station 4 and the wide area communication network 3 at a prescribed period. In addition to the source information indicating the vehicle that sent the communication data packet (i.e., the source vehicle), the vehicle status report also includes the time when the data was generated, the current position of the source vehicle, etc. The source information is identification information (the so-called vehicle ID) pre-assigned to the source vehicle to distinguish it from other vehicles. In addition to the above information, the vehicle status report may also include the direction of travel of the vehicle, the driving lane ID, the driving speed, the acceleration, the yaw rate, etc. The driving lane ID indicates which lane the vehicle is traveling in from the left or right end of the road. In addition, the vehicle status report may also include information such as the lighting status of the direction indicator and whether the vehicle is traveling across the lane dividing line.

[0054] Furthermore, each in-vehicle system 1 uploads a communication data packet (hereinafter referred to as a roadside parking location report) indicating information related to roadside parking locations notified from the map server 2 to the map server 2. Information related to roadside parking locations is information used by the map server 2 to determine the presence or absence of roadside parked vehicles. The roadside parking location report may be included in the vehicle status report, or the roadside parking location report and vehicle status report may be sent separately.

[0055] The map server 2 detects the locations of parked vehicles and their disappearance based on the parked location reports uploaded by each vehicle, and then multicasts information on the presence / disappearance of parked vehicles to the vehicles to which the information should be distributed.

[0056] The map server 2 has a function of managing the current position of each vehicle as a sub-function for determining the distribution destination of information about the appearance / disappearance of roadside parked vehicles. The management of the current position of each vehicle can be achieved using a prescribed database, namely a vehicle position database. The current position of each vehicle is stored in the database in a manner that establishes a correspondence with the vehicle ID, etc. Whenever the map server 2 receives a vehicle status report, it refers to the content and updates the current position of the sending source vehicle registered in the database. In addition, in the structure of pull-type distribution of roadside parking location information, for example, a structure such as a vehicle position database for determining the distribution destination of roadside parking location information is not necessarily required. The function of managing the position of each vehicle for determining the distribution destination is an arbitrary element. The transmission of the vehicle status report in the vehicle-mounted system 1 is also an arbitrary element.

[0057] <Overview of In-Vehicle System 1>

[0058] Figure 2 The in-vehicle system 1 shown includes a front camera 11, a millimeter-wave radar 12, a vehicle status sensor 13, a positioner 14, a V2X on-board device 15, an HMI system 16, a map collaboration device 50, and a driver assistance ECU 60. The term "ECU" in the component names is an abbreviation for "Electronic Control Unit," meaning an electronic control unit. Furthermore, "HMI" is an abbreviation for "Human Machine Interface." V2X, an abbreviation for "Vehicle to X (Everything)," refers to the communication technology that connects the vehicle to various objects.

[0059] The various devices or sensors constituting the vehicle-mounted system 1 are connected as nodes to the communication network constructed in the vehicle, namely the in-vehicle network Nw. The nodes connected to the in-vehicle network Nw can communicate with each other. In addition, specific devices can also be configured to communicate directly with each other without going through the in-vehicle network Nw. For example, the map cooperation device 50 and the driving assistance ECU 60 can also be directly electrically connected through a dedicated line. In addition, although Figure 2 The in-vehicle network Nw is configured as a bus, but is not limited to this. The network topology may also be a mesh, star, or ring topology. The network configuration can be modified as appropriate. Various standards, such as the Controller Area Network (hereinafter referred to as CAN: a registered trademark), Ethernet (Ethernet is a registered trademark), and FlexRay (registered trademark), can be adopted as standards for the in-vehicle network Nw.

[0060] Hereinafter, the passenger seated in the driver's seat of this vehicle, also known as the driver's seat passenger, will also be referred to as the user. Furthermore, the front-to-back, left-to-right, and up-to-down directions in the following description are defined with respect to this vehicle. Specifically, the front-to-back direction corresponds to the longitudinal direction of this vehicle. The left-to-right direction corresponds to the width of this vehicle. The up-to-down direction corresponds to the height of the vehicle. From another perspective, the up-to-down direction corresponds to a direction perpendicular to a plane parallel to the front-to-back and left-to-right directions.

[0061] Components of the In-Vehicle System 1

[0062] The front camera 11 is a camera that captures the front of the vehicle at a predetermined viewing angle. The front camera 11 is, for example, arranged at the upper end of the front glass on the inner side of the vehicle cabin, or at the front grille, roof, etc. The front camera 11 includes a camera body that generates image frames, and an ECU that detects predetermined detection objects by performing recognition processing on the image frames. The camera body is composed of at least an image sensor and a lens, and generates and outputs captured image data at a predetermined frame rate (for example, 60fps). The camera ECU is mainly composed of an image processing chip including a CPU or GPU, and includes a recognizer as a functional module. The recognizer recognizes the type of object based on, for example, a feature vector of the image.

[0063] The front camera 11 detects a specified detection object and determines the relative position of the detection object relative to the vehicle. The detection object here refers to, for example, pedestrians, other vehicles, landmark features, road ends, road surface markings, etc. Other vehicles also include bicycles, bicycles with prime movers, and motorcycles. Landmarks are three-dimensional structures set up along the road. Structures set up along the road include, for example, guardrails, curbs, trees, telephone poles, road signs, traffic lights, etc. Road signs include guide signs such as direction signs and road name signs. Landmark features are used for the localization processing described later. Road surface markings refer to spray patterns drawn on the road surface for traffic control and traffic restrictions. For example, lane dividing lines (so-called lane markings) indicating the boundaries of lanes, crosswalks, stop lines, guide strips, safety zones, restriction arrows, etc. are included in road surface markings. Lane dividing lines, in addition to spray patterns formed as dotted lines or continuous lines using yellow or white paint, also include lane dividing lines implemented by road studs such as Chatter bars or Botts' Dots. Lane dividing lines are also called lane markings or lane markings.

[0064] Furthermore, when the front camera 11 detects a vehicle, it determines whether it is a parked vehicle (a stopped vehicle) based on the relative speed of the detected vehicle. Furthermore, the front camera 11 outputs information indicating, for example, the position of the parked vehicle on the road or its relative position relative to the host vehicle. Furthermore, the front camera 11 is preferably configured to detect not only the lane in which the host vehicle is traveling, but also parked vehicles in areas corresponding to adjacent lanes. As an example, the front camera 11 is configured to detect vehicles in the host vehicle's lane and in the adjacent lanes to the left and right.

[0065] The image processor of the front camera 11 separates and extracts the background and the detection object from the captured image based on image information including color, brightness, and contrast related to color and brightness. For example, the front camera 11 uses SfM (Structure from Motion) processing and the like to calculate the relative distance and direction (that is, relative position) and moving speed of the detection object such as the lane dividing line, the road end, and the vehicle from the vehicle. The relative position of the detection object relative to the vehicle can also be determined based on the size and inclination of the detection object in the image. In addition, the detection result data indicating the position, type, etc. of the detection object are sequentially provided to the map cooperation device 50 and the driving assistance ECU 60.

[0066] The millimeter-wave radar 12 is a device that detects the relative position and relative speed of an object relative to the vehicle by sending millimeter waves or quasi-millimeter waves toward the front of the vehicle and analyzing the received data of the reflected wave when the transmitted wave is reflected by the object and returns. The millimeter-wave radar 12 is, for example, set in the front grille or front bumper. The millimeter-wave radar 12 has a built-in radar ECU that identifies the type of the detected object based on the size, moving speed, and receiving intensity of the detected object. The radar ECU outputs data indicating the type, relative position (direction and distance), and receiving intensity of the detected object to the map cooperation device 50, etc. as a detection result. The millimeter-wave radar 12 is also configured to be able to detect part or all of the above-mentioned roadside parked vehicles. For example, the millimeter-wave radar 12 detects parked vehicles based on the position, moving speed, size, and reflection intensity of the detected object.

[0067] The front camera 11 and the millimeter-wave radar 12 can also be configured to provide observation data used for object recognition, such as image data, to the driving assistance ECU 60, etc. via the in-vehicle network Nw, in addition to data representing the recognition results. For example, the observation data for the front camera 11 refers to image frames. The observation data of the millimeter-wave radar refers to data representing the receiving intensity and relative speed of each detection direction and distance, or data representing the relative position and receiving intensity of the detected object. The observation data is equivalent to the raw data observed by the sensor, or the data before the recognition processing is performed. In addition, the front camera 11 and the millimeter-wave radar 12 are both equivalent to sensors that sense the outside world of the vehicle. Therefore, without distinguishing between the front camera 11 and the millimeter-wave radar 12, they are also recorded as peripheral monitoring sensors.

[0068] Object recognition processing based on observation data generated by the surrounding monitoring sensors can also be performed by an ECU other than the sensor, such as the driving assistance ECU 60. A portion of the functions of the front camera 11 and the millimeter-wave radar 12 can also be incorporated into the driving assistance ECU 60. In this case, the camera serving as the front camera 11 and the millimeter-wave radar only need to provide observation data such as image data and distance measurement data to the driving assistance ECU 60 as detection result data.

[0069] The vehicle state sensor 13 is a sensor that detects state quantities related to the driving control of the vehicle. The vehicle state sensor 13 includes, for example, inertial sensors such as a three-axis gyroscope sensor and a three-axis acceleration sensor. The three-axis acceleration sensor is a sensor that detects the accelerations acting on the vehicle in the front-back, left-right, and up-down directions. The gyroscope sensor is a sensor that detects the angular velocity of rotation around the detection axis. The three-axis gyroscope sensor refers to a sensor with three mutually orthogonal detection axes. In addition, the vehicle state sensor 13 can also include a gear position sensor, a steering angle sensor, a vehicle speed sensor, etc. The gear position sensor is a sensor that detects the position of the gear lever. The steering angle sensor is a sensor that detects the rotation angle of the steering wheel (the so-called steering angle). The vehicle speed sensor is a sensor that detects the driving speed of the vehicle.

[0070] The vehicle state sensor 13 outputs data representing the current value of the physical state quantity being detected (i.e., the detection result) to the in-vehicle network Nw. The output data of each vehicle state sensor 13 is obtained by the map cooperation device 50 and the like via the in-vehicle network Nw. The types of sensors used by the in-vehicle system 1 as the vehicle state sensors 13 can be appropriately designed, and it is not necessary to include all of the sensors described above.

[0071] The locator 14 is a device that generates high-precision position information of the vehicle by combining multiple information for composite positioning. Figure 3As shown, the locator 14 is implemented using a GNSS receiver 141 , an inertial sensor 142 , a map storage unit 143 , and a position calculation unit 144 .

[0072] The GNSS receiver 141 sequentially detects its current position by receiving navigation signals transmitted from positioning satellites that make up the GNSS (Global Navigation Satellite System). For example, if the GNSS receiver 141 can receive navigation signals from four or more positioning satellites, it outputs positioning results every 100 milliseconds. Examples of GNSS systems include GPS, GLONASS, Galileo, IRNSS, QZSS, and Beidou. Examples of inertial sensors 142 include a three-axis gyroscope and a three-axis accelerometer.

[0073] The map storage unit 143 is a non-volatile memory that stores high-precision map data. High-precision map data here refers to map data that shows the road structure and the coordinates of features along the road with accuracy sufficient for autonomous driving. High-precision map data includes, for example, three-dimensional road shape data, lane data, and feature data. The three-dimensional road shape data includes node data related to locations where multiple roads intersect, merge, or branch (hereinafter referred to as nodes), and link data related to the roads connecting these locations (hereinafter referred to as links). Link data may also include data indicating the road type, such as whether it is a dedicated road or a general road. Dedicated roads here refer to roads where pedestrians and bicycles are prohibited, such as toll roads like expressways. Road type may also include attribute information indicating whether autonomous driving is permitted. Lane data indicates the number of lanes, the coordinates of lane dividing lines (so-called lane markings), the direction of travel for each lane, and the locations where lanes branch or merge. Feature data includes the location and type of road markings such as stop signs, as well as the location, shape, and type of landmarks. Landmarks include three-dimensional structures installed along roads, such as traffic signs, signal lights, poles, and commercial signs.

[0074] The position calculation unit 144 sequentially measures the vehicle's position by combining the positioning results from the GNSS receiver 141 with the measurement results from the inertial sensor 142. For example, in situations where the GNSS receiver 141 cannot receive GNSS signals, such as in a tunnel, the position calculation unit 144 performs dead reckoning (autonomous navigation) using the yaw rate and vehicle speed. The yaw rate used for dead reckoning can be either the yaw rate calculated by the front camera 11 using the SfM technique or the yaw rate detected by the yaw rate sensor. The measured vehicle position information is output to the in-vehicle network Nw and utilized by the map cooperation device 50, etc. Furthermore, based on the vehicle position coordinates determined by the above-described configuration, the position calculation unit 144 determines the ID of the lane on the road in which the vehicle is traveling (hereinafter referred to as the driving lane).

[0075] In addition, the locator 14 can also be configured to be able to perform positioning processing. Positioning processing refers to the process of determining the detailed position of the vehicle by comparing the coordinates of the landmark determined based on the image captured by the front camera 11 with the coordinates of the landmark registered in the high-precision map data. Positioning processing can also be implemented by comparing the three-dimensional detection point group data output by LiDAR (Light Detection and Ranging / Laser Imaging Detection and Ranging) with the three-dimensional map data. In addition, the locator 14 can also be configured to determine the driving lane based on the distance from the road end detected by the front camera 11 and the millimeter wave radar 12. The map cooperation device 50 or the driving assistance ECU 60 can also have some or all of the functions of the locator 14.

[0076] The V2X vehicle-mounted device 15 is used to wirelessly communicate between the vehicle and other devices. The "V" in V2X refers to the vehicle itself, and the "X" refers to various entities outside the vehicle, such as pedestrians, other vehicles, road equipment, networks, and servers. The V2X vehicle-mounted device 15 includes a wide-area communication unit and a narrow-area communication unit as communication modules. The wide-area communication unit is used to conduct wireless communications based on a prescribed wide-area wireless communication standard. Examples of such standards include LTE (Long Term Evolution), 4G, and 5G. Furthermore, the wide-area communication unit can be configured to communicate directly with other devices, that is, without a base station, in addition to communicating via wireless base stations. In other words, the wide-area communication unit can be configured to implement cellular V2X. The installation of the V2X vehicle-mounted device 15 makes the vehicle a connected car capable of connecting to the network. For example, the map cooperation device 50 can download the latest high-precision map data from the map server 2 by cooperating with the V2X vehicle-mounted device 15 and update the map data stored in the map storage unit 143 .

[0077] The narrow area communication unit of the V2X vehicle-mounted device 15 is a communication module for directly implementing wireless communication with other mobile bodies and roadside devices that exist around the vehicle through a communication standard (hereinafter referred to as a narrow area communication standard) whose communication distance is limited to several hundred meters. As other mobile bodies, they are not limited to vehicles, but can include pedestrians, bicycles, etc. As narrow area communication standards, any standard such as the WAVE (Wireless Access in Vehicular Environment) standard disclosed by IEEE1709 and the DSRC (Dedicated Short Range Communications) standard can be adopted. The narrow area communication unit broadcasts the vehicle information of the vehicle to the surrounding vehicles at a prescribed transmission period, and receives vehicle information sent from other vehicles. Vehicle information includes vehicle ID, current location, direction of travel, moving speed, operation status of the direction indicator, timestamp, etc.

[0078] The HMI system 16 provides an input interface function for accepting user operations and an output interface function for presenting information to the user. The HMI system 16 includes a display 161 and an HCU (HMI Control Unit) 162. In addition to the display 161, other means for presenting information to the user can include a speaker, a vibrator, or a lighting device (e.g., an LED).

[0079] Display 161 is a device that displays images. For example, display 161 is a so-called center display located at the topmost portion of the center section of the instrument panel in the vehicle width direction (hereinafter referred to as the center area). Display 161 is capable of full-color display and can be implemented using a liquid crystal display, an OLED (Organic Light Emitting Diode) display, a plasma display, or the like. Alternatively, the HMI system 16 may include a head-up display (HUD) that projects a virtual image onto the portion of the windshield in front of the driver's seat. Alternatively, display 161 may be an instrument display.

[0080] The HCU 162 is a component that controls the information presented to the user. The HCU 162 is implemented using a processor such as a CPU or GPU, RAM, and flash memory. The HCU 162 controls the display screen of the display 161 based on information provided by the map cooperation device 50 and signals from an input device (not shown). For example, the HCU 162 displays the information on the display 161 based on a request from the map cooperation device 50 or the driving assistance ECU 60. Figure 4 The illustrated roadside parked vehicle notification image 80 .

[0081] The parked vehicle notification image 80 is an image used to notify the user of information related to parked vehicles. Preferably, the parked vehicle notification image 80 includes information such as the distance to the parked vehicle, characteristics of the parked vehicle (such as vehicle type and color), and the degree of protrusion into the lane. Figure 4 Image 81 in the image represents the vehicle, and image 82 represents the lane dividing line. Image 83 represents a parked vehicle, and image 84 represents the road edge. The color tone of image 83 representing the parked vehicle is preferably similar to the color of the actual parked vehicle. Furthermore, the parked vehicle notification image 80 may also include image 85 indicating the remaining distance to the location of the parked vehicle. The parked vehicle notification image 80, which indicates the location of the parked vehicle, may also be displayed on the head-up display superimposed on the real world as viewed from the driver's seat.

[0082] The map collaboration device 50 is a device that obtains map data containing roadside parking location information from the map server 2 and uploads information about roadside parked vehicles detected by the vehicle to the map server 2. Details of the functions of the map collaboration device 50 will be described later. The map collaboration device 50 is primarily composed of a computer comprising a processing unit 51, RAM 52, a storage device 53, a communication interface 54, and a bus connecting these. The processing unit 51 is hardware that is integrated with the RAM 52 for computational processing. The processing unit 51 includes a computing core such as at least one CPU (Central Processing Unit). The processing unit 51 accesses the RAM 52 to execute various processes for determining the presence or absence of roadside parked vehicles. The storage device 53 includes a non-volatile storage medium such as flash memory. The storage device 53 stores a program executed by the processing unit 51, namely, a roadside parked vehicle reporting program. Execution of the roadside parked vehicle reporting program by the processing unit 51 is equivalent to executing the parking location management method corresponding to the roadside parked vehicle reporting program. The communication interface 54 is a circuit for communicating with other devices via the in-vehicle network Nw. The communication interface 54 may be realized using analog circuit elements, ICs, and the like.

[0083] Furthermore, the map coordination device 50 may be included in, for example, a navigation device. It may also be included in the driving assistance ECU 60 or the autonomous driving ECU. It may also be included in the V2X vehicle-mounted device 15. The functional configuration of the map coordination device 50 can be modified as appropriate. The map coordination device 50 corresponds to a vehicle device.

[0084] The driving assistance ECU 60 assists the driver's seat passenger in driving operations based on detection results from surrounding monitoring sensors such as the front camera 11 and millimeter-wave radar 12, and map information acquired by the map collaboration device 50. For example, the driving assistance ECU 60 displays driving assistance information such as a parked vehicle notification image showing the location of parked vehicles. Furthermore, the driving assistance ECU 60 controls driving actuators based on detection results from the surrounding monitoring sensors and map information acquired by the map collaboration device 50, thereby performing some or all driving operations on behalf of the driver's seat passenger. Examples of driving actuators include brake actuators, electronic throttle valves, and steering actuators.

[0085] The driving assistance ECU 60 provides a function for automatically executing lane changes (hereinafter referred to as the automatic lane change function) as one of the vehicle control functions. For example, upon reaching a predetermined lane change location on a separately generated driving plan, the driving assistance ECU 60, in coordination with the HMI system 16, queries the driver's seat passenger whether to execute a lane change. Furthermore, if it is determined that the driver's seat passenger has input a lane change instruction, the ECU applies a steering force toward the target lane, taking into account the traffic conditions in the target lane, to shift the vehicle's driving position into the target lane. The predetermined lane change location can be defined as an interval of a certain length.

[0086] The driving assistance ECU 60, like the map cooperation device 50, is primarily composed of a computer including a processing unit, RAM, a storage device, a communication interface, and a bus connecting these components. Illustration of these components is omitted. The storage device included in the driving assistance ECU 60 stores a program executed by the processing unit, namely, a driving assistance program. The execution of the driving assistance program by the processing unit is equivalent to executing the method corresponding to the driving assistance program.

[0087] <Details of Map Cooperation Device 50>

[0088] Here we use Figure 5 The functions and operations of the map cooperation device 50 are described. The map cooperation device 50 provides the information related to the roadside parked vehicle reporting program stored in the storage device 53. Figure 5 The various functional modules shown in the figure correspond to their functions. Specifically, the map coordination device 50 includes a vehicle position acquisition unit F1, a map acquisition unit F2, a vehicle behavior acquisition unit F3, a detected object information acquisition unit F4, a report data generation unit F5, and a notification processing unit F6 as functional modules. The map acquisition unit F2 includes a roadside parking information acquisition unit F21. The report data generation unit F5 includes a roadside parking presence determination unit F51.

[0089] The vehicle position acquisition unit F1 acquires the position information of the vehicle from the locator 14. It also acquires the driving lane ID from the locator 14. The vehicle position acquisition unit F1 may have a part or all of the functions of the locator 14.

[0090] The map acquisition unit F2 reads map data for a specified range based on the current location from the map storage unit 143. Furthermore, the map acquisition unit F2 obtains information on roadside parking locations within a specified distance ahead of the vehicle from the map server 2 via the V2X vehicle-mounted device 15. As will be described later, this roadside parking location information is data regarding the locations of parked vehicles and includes information such as the location coordinates of the parked vehicles and the types of the parked vehicles. The component that obtains roadside parking location information from the map server 2 corresponds to the roadside parking information acquisition unit F21.

[0091] The roadside parking location information acquisition unit F21 can obtain roadside parking location information by requesting the map server 2 for roadside parking location information corresponding to the vehicle's location. This distribution method is also known as pull distribution. Alternatively, the map server 2 can automatically distribute roadside parking location information to vehicles located near parked vehicles. This distribution method is also known as push distribution. In other words, roadside parking location information can be acquired using either pull or push distribution. As an example, the map server 2 selects vehicles to be distributed based on the location information of each vehicle and performs push distribution on these vehicles.

[0092] The roadside parking location information acquired by the map acquisition unit F2 is temporarily stored in a memory M1 implemented using, for example, the RAM 52. The roadside parking location information stored in the memory M1 is deleted when the vehicle passes the location indicated by the data or after a predetermined time has elapsed. For convenience, the roadside parking location information acquired from the map server 2 is also recorded as on-map roadside parking location information.

[0093] The vehicle behavior acquisition unit F3 acquires data representing the vehicle's behavior from the vehicle state sensor 13. For example, it acquires data such as driving speed, yaw rate, lateral acceleration, and longitudinal acceleration. Furthermore, the vehicle behavior acquisition unit F3 acquires information indicating whether the vehicle has crossed a lane boundary and the rightward or leftward offset of the vehicle's driving position relative to the lane center from the front camera 11. Longitudinal acceleration here corresponds to acceleration in the front-to-back direction, and lateral acceleration corresponds to acceleration in the left-to-right direction.

[0094] The detection object information acquisition unit F4 acquires information about parked vehicles detected by the front camera 11 and the millimeter wave radar 12 (hereinafter referred to as parked vehicle detection information). The detection object information acquisition unit F4 is equivalent to the parked vehicle information acquisition unit. The parked vehicle detection information includes, for example, the location of the parked vehicle, its type, size, etc. For example, the detection position of the parked vehicle can be expressed in any absolute coordinate system such as WGS84 (World Geodetic System 1984: 1984 World Geodetic Coordinate System). The absolute position of the detected parked vehicle can be calculated by combining the current position coordinates of the vehicle and the relative position information of the parked vehicles detected by the surrounding monitoring sensors relative to the vehicle. The detection object information acquisition unit F4 can not only acquire the recognition results of various surrounding monitoring sensors, but also acquire the observation data itself, such as the image data captured by the front camera 11. The detection object information acquisition unit F4 can also be called an external information acquisition unit.

[0095] The detected position information of a parked vehicle preferably includes information about the distance from the road end to the parked vehicle and the relative position of the parked vehicle with respect to the lane boundary. For example, information indicating whether the parked vehicle protrudes into the lane and the degree of protrusion into the lane may also be included. Alternatively, the lateral position of the end of the parked vehicle within the lane may also be included. If the parked vehicle is parked on the roadside, this lateral position information of the end of the parked vehicle within the lane can be used to indicate how much of the lane is blocked by the parked vehicle.

[0096] The various data acquired in sequence by the vehicle position acquisition unit F1, the vehicle behavior acquisition unit F3, and the detection object information acquisition unit F4 are stored in a memory such as RAM52, and are referenced and utilized by the map acquisition unit F2, the report data generation unit F5, etc. In addition, various information is divided according to type and stored in the memory after, for example, a timestamp indicating the acquisition time of the data is attached. The timestamp plays the role of establishing an association between different types of information at the same time. By using the timestamp, the map cooperation device 50 can, for example, determine the vehicle behavior synchronized with the video outside the vehicle. In addition, the timestamp can also replace the acquisition time and be the output time, generation time, etc. of the data in the output source. When the output time or generation time is used as the timestamp, it is preferred to synchronize the time information of each vehicle-mounted device. The various information acquired by the map cooperation device 50 can be sorted and stored in a manner such that the latest data is at the beginning. Data that has passed a constant time since acquisition can be discarded.

[0097] The report data generator F5 generates a data set to be sent to the map server 2 and outputs it to the V2X vehicle-mounted device 15. For example, the report data generator F5 generates the vehicle status reports described above at predetermined intervals and uploads them to the map server 2 via the V2X vehicle-mounted device 15. Furthermore, the report data generator F5 generates roadside parking location reports and uploads them to the map server 2 as part of the upload process described later. The report data generator F5 functions as a report processing unit.

[0098] The curbside parking presence determination unit F51 determines whether a vehicle is parked on the curb based on the parked vehicle detection information acquired by the detected object information acquisition unit F4 and the vehicle's behavior data acquired by the vehicle behavior acquisition unit F3. For example, if a parked vehicle detected by the perimeter monitoring sensor meets the curbside parking conditions, the curbside parking presence determination unit F51 considers the parked vehicle to be a curbside vehicle. The curbside parking conditions distinguish between vehicles parked temporarily within a lane, such as those due to waiting for a traffic light or traffic congestion, and vehicles parked close to the roadside shoulder.

[0099] For example, as roadside parking conditions, (1) there is no intersection or traffic light within a first distance specified on the road travel direction side from the parked vehicle, and (2) the distance from the road end to the parked vehicle is less than a specified second distance. The first distance can be, for example, 10m. The second distance can be, for example, 0.75m. The positional relationship with the intersection can be determined by referring to map data. In addition, the lateral distance from the road end to the parked vehicle can be determined based on the detection results of the surrounding monitoring sensors such as the front camera 11. In addition, the roadside parking presence determination unit F51 can also use map data to determine the distance from the road end to the parked vehicle. For example, the roadside parking presence determination unit F51 determines the position of the parked vehicle on the map by combining the vehicle position determined as a result of the positioning process and the relative position of the parked vehicle relative to the vehicle obtained by the detection object information acquisition unit F4. Moreover, the lateral distance from the road end to the parked vehicle can also be determined based on the position information of the road end shown in the map data and the position information of the parked vehicle.

[0100] The roadside parking condition may also include (3) not being a parking prohibited section. In addition, (4) the parked vehicle has the hazard lights on, (5) the brake lights are not on, (6) the door of the parked vehicle is open, (7) crossing the lane dividing line, etc. may be used as materials for determining whether the vehicle is a roadside parked vehicle. In addition, in order to distinguish it from traffic congestion waiting to turn left or right, (8) when multiple parked vehicles are in a row, the row is not connected to the intersection, etc. can also be included. The parking prohibited section can be determined based on the recognition results of the front camera 11 on road signs, road surface display, the color of the curb at the end of the road, etc. Map data can also be used to determine whether it is a roadside parked section. The lighting status of the hazard lights and brake lights and the open / close status of the doors can be determined based on the recognition results of the front camera 11. In addition, the roadside parking condition does not need to include all of (1) to (8). Part of (1) to (8) or viewpoints other than the above can be used as conditions for determining whether the vehicle is a roadside parked vehicle.

[0101] The roadside parking presence determination unit F51 is configured to determine, on one side, whether a vehicle stopped on the road, detected based on input signals from the surrounding monitoring sensor, is a roadside parked vehicle or a temporarily parked vehicle. The roadside parking presence determination unit F51 corresponds to a parking determination unit.

[0102] Alternatively, the roadside parking presence determination unit F51 may determine whether there is a parked vehicle based on whether the vehicle takes evasive action to avoid a parked vehicle detected by the perimeter monitoring sensor near the vehicle's driving lane.

[0103] The avoidance action here is, for example, a vehicle movement to avoid a parked vehicle on the roadside, such as a change in driving position. The change in driving position here refers to a change in the lateral position of the vehicle on the road. The change in driving position includes not only lane changes, but also movement of the driving position within the same lane to either the left or right corner, and driving across lane dividing lines. In addition, in order to clarify the difference from ordinary lane changes, the avoidance action is preferably a change in driving position / steering operation accompanied by deceleration and subsequent acceleration. For example, a change in driving position accompanied by a deceleration operation, or a change in driving position accompanied by deceleration to a speed below a specified speed can be considered as an avoidance action. In addition, the above description of the avoidance action shows the concept of the avoidance action assumed in the present disclosure. As will be described later, in addition to the driving trajectory, it is possible to determine whether a change in driving position as an avoidance action has been performed based on the change pattern of lateral acceleration, the movement history of the direction indicator, etc.

[0104] When a parked vehicle detected by the peripheral monitoring sensor is determined to be a parked vehicle, the roadside parking presence determination unit F51 stores the parked vehicle detection information acquired by the detection object information acquisition unit F4 as roadside parked vehicle detection information. The roadside parked vehicle detection information has the same data structure as the parked vehicle detection information, for example, including the location of the roadside parked vehicle, its vehicle type, etc. The roadside parked vehicle detection information may also include the color of the roadside parked vehicle, the distance from the road end to the roadside parked vehicle, and the relative position information of the roadside parked vehicle with respect to the lane dividing line. Furthermore, the roadside parked vehicle detection information may also include image data of the roadside parked vehicle captured by the front camera 11. At least one of the roadside parked vehicle detection information and the parked vehicle detection information corresponds to the sensing information of the peripheral monitoring sensor.

[0105] The notification processing unit F6 is configured to cooperate with the HMI system 16 to notify the driver's seat passenger of information about a vehicle parked on the roadside in front of the vehicle based on the roadside parking information on the map. For example, the notification processing unit F6 generates a Figure 4 The illustrated roadside parked vehicle notification image is displayed on the display 161. In addition, the roadside parked vehicle notification may be made by a voice message, etc. The driving support ECU 60 may also include a notification processing unit F6.

[0106] Upload Processing

[0107] Here we use Figure 6 The flowchart shown in the figure explains the upload process executed by the map cooperation device 50. For example, while the vehicle is powered on, the upload process is executed at a predetermined period such as every 100 milliseconds. Figure 6 The driving power supply is the power supply that enables the vehicle to travel, for example, the ignition power supply in an engine vehicle. In electric vehicles, the system main relay corresponds to the driving power supply. The upload process includes steps S101 to S104 as an example.

[0108] In step S101, the report data generator F5 reads the roadside parking location information on the map stored in the memory M1 and moves to step S102. In step S102, based on the roadside parking location information on the map, it is determined whether there is a roadside parked vehicle within a specified distance (hereinafter referred to as the reference distance) in front of the vehicle. The reference distance is, for example, 200m or 300m. Preferably, the reference distance is longer than the limit value of the distance at which the front camera 11 can recognize an object. The reference distance can also be changed according to the vehicle's driving speed. For example, the reference distance can be set to be longer as the vehicle's driving speed increases. For example, the distance reached within a specified time such as 30 seconds can be calculated based on the vehicle's speed and used as the reference distance.

[0109] If there is no roadside parked vehicle recognized by the map server 2 within the reference distance in step S102, this flow ends. On the other hand, if there is a roadside parked vehicle within the reference distance, step S103 is executed.

[0110] In step S103, the vehicle's behavior is acquired when traveling within a specified reporting distance before and after a roadside stop on the map, and the process proceeds to step S104. In step S104, a dataset is generated as a roadside stop report, including the time-series data on vehicle behavior acquired in step S103, transmission source information, and report target location information. The report target location information indicates the location for which the report is being made. For example, the report target location information includes the coordinates of the roadside stop on the map.

[0111] The reporting target distance is preferably set to a distance at which the driver's seat passenger and the surrounding monitoring sensor can recognize the status of the roadside parking location on the map. Figure 7 The setting shown is 100 meters before and after the roadside stop on the map. In this case, the roadside stop report will be a dataset showing vehicle behavior within 100 meters before and after the roadside stop on the map. The sections within the report target distance before and after the roadside stop on the map are also recorded as report target sections.

[0112] The vehicle behavior data included in the roadside stop report indicates whether the vehicle, traveling in a lane with a parked vehicle, has taken maneuvers to avoid the parked vehicle (i.e., avoidance action). For example, data indicating vehicle behavior can include the vehicle's position coordinates, travel direction, travel speed, longitudinal acceleration, lateral acceleration, and yaw rate at each moment of passage near the roadside stop on a map. The vicinity of a roadside stop on a map, for example, refers to within 20 meters of the roadside stop on a map. Alternatively, the vicinity of the roadside stop on a map can be defined as within 50 meters or 100 meters before and after the roadside stop on a map. The range considered near the roadside stop on a map can also be modified based on the road type or the legal speed limit. The distance considered near the roadside stop on a map determines the distance covered by the report. Furthermore, data indicating vehicle behavior can include steering angle, gear position, turn signal status, hazard light illumination, whether a lane dividing line has been crossed, whether a lane change has been made, and the amount of deviation from the lane center.

[0113] The roadside parking location report preferably includes the lane ID at each time the vehicle passed near the roadside parking location on the map. This is because the inclusion of the lane ID allows the map server 2 to determine whether the report is from a vehicle traveling in the lane affected by the roadside parking location (i.e., the first lane). Of course, the map server 2 can also determine whether the report is from a vehicle traveling in the lane affected by the roadside parking location based on the time series data of the position coordinates included in the roadside parking location report.

[0114] Furthermore, it is preferable that the roadside parking location report include not only vehicle behavior up to the point of arrival at the mapped roadside parking location, but also vehicle behavior after passing the mapped roadside parking location. This is because if a vehicle changes lanes or maneuvers to avoid a parked vehicle, there is a high probability that the vehicle will return to its original lane after passing the parked vehicle. In other words, by also including vehicle behavior after passing the mapped roadside parking location in the roadside parking location report, the accuracy of determining whether the vehicle's maneuver was to avoid a parked vehicle can be improved, thereby improving the accuracy of determining whether a parked vehicle actually exists.

[0115] Roadside parking location reports can be data indicating the vehicle's status every 100 milliseconds, for example, while traveling within the reporting area. The sampling interval for vehicle behavior is not limited to 100 milliseconds and can also be 200 milliseconds, for example. A shorter sampling interval increases the data size, so from the perspective of traffic control, it is preferable to set the sampling interval long enough to analyze vehicle behavior.

[0116] If the reporting target distance is too short, for example, the map server 2 will only collect data after evasive action has been taken, making it unclear whether evasive action was taken. On the other hand, setting the reporting target distance longer reduces the omission of data indicating evasive action, but increases the data size. It is preferable to set the reporting target distance to include the location where evasive action is assumed to be taken against a parked vehicle. For example, it is preferable to set the reporting target distance to at least 25 meters.

[0117] Furthermore, the length of the reporting distance can be changed depending on whether the road is a general road or a dedicated road for motorists. Dedicated roads for motorists are roads where pedestrians and bicycles are prohibited from entering, including toll roads such as expressways. For example, the reporting distance on general roads can be set shorter than that on dedicated roads for motorists. Specifically, the reporting distance for dedicated roads can be set to 100 meters or longer, while the reporting distance for general roads can be set to 30 meters or shorter, for example, to 50 meters or shorter. This is because dedicated roads offer better visibility ahead than general roads, making it possible to initiate evasive action from a location farther away from parked vehicles on the roadside.

[0118] The sampling interval can also be changed depending on the road type, such as whether it is a dedicated road or a general road. The sampling interval for dedicated roads can also be shorter than that for general roads. Extending the sampling interval can reduce data size. Furthermore, the sampling interval can be configured to be less frequent as the distance to be reported increases. Such a configuration can keep the data size of roadside stop reports within a constant range.

[0119] Furthermore, the report target distance and sampling interval may be dynamically determined based on an instruction signal from the map server 2. Furthermore, the types of information (ie, items) included in the roadside stop report may also be dynamically determined based on an instruction signal from the map server 2.

[0120] Furthermore, the reporting distance, sampling interval, and information included in the roadside parking location report can be modified based on the type and size of parked vehicles, as well as the degree of lane congestion. For example, in scenarios where a lane change is necessary as an evasive maneuver, such as when a parked vehicle blocks more than half of a lane, the roadside parking location report can be limited to information used to determine whether the reporting vehicle has executed a lane change. Lane change determination can be based on factors such as the driving trajectory and the presence or absence of changes in the lane ID.

[0121] In addition, the roadside parking location report may include detection result information indicating whether the surrounding monitoring sensor has detected a roadside parked vehicle. The roadside parked vehicle detection result can be the detection result of the front camera 11 and the millimeter wave radar 12 respectively. The roadside parked vehicle detection result can also be the determination result of the roadside parking presence determination unit F51. In the case of detecting a roadside parked vehicle, the roadside parking location report may include at least one of the roadside parked vehicle detection information and the parked vehicle detection information as the sensing information of the surrounding monitoring sensor. For example, the roadside parking location report may include image data captured by the front camera 11 at a specified distance from the roadside parking location on the map (for example, 10 meters).

[0122] In addition, the upload processing method is not limited to the above content. Figure 8 As shown, the process may also be configured to include steps S201 to S206. Figure 8 The steps S201 to S203 are the same as the above steps S101 to S101. If step S203 is completed, step S204 is executed.

[0123] In step S204, sensing information from at least one of the front camera 11 and the millimeter-wave radar 12 is acquired when the vehicle passes near a roadside parking spot on the map. This sensing information may include not only recognition results based on observation data, but also the observation data itself. As an example, the vehicle acquires recognition results related to parked vehicles from the front camera 11 and the millimeter-wave radar 12, as well as images captured by the front camera 11. The parked vehicles here represent potential roadside parking vehicles. The sensing information collection period, similar to vehicle behavior information, can be from the time the vehicle passes a point where the remaining distance to the roadside parking spot on the map becomes less than the reporting distance until the roadside parking spot on the map is positioned behind the reporting distance. Alternatively, if a perimeter monitoring sensor with a detection range behind the vehicle is not present, the sensing information collection period may also be from the time the remaining distance to the roadside parking spot on the map becomes less than the reporting distance until the vehicle passes the roadside parking spot on the map. Once step S204 is completed, step S205 is executed.

[0124] In step S205, current status data representing the current status of the roadside parking location on the map is generated based on the sensing information collected in step S204. For example, the current status data includes the recognition results of the surrounding monitoring sensors every 250 milliseconds during the sensing information collection period. Furthermore, if a parked vehicle was detected by the front camera 11 during this period, the data also includes at least one frame of image data used for detecting the parked vehicle. By including at least one image frame showing the roadside parking location on the map in the current status data, the resolution of the map server 2 can be improved.

[0125] Furthermore, the image frames included in the current situation data may include all frames captured during the period of sensing information collection, or image frames captured at 200 millisecond intervals. The greater the number of image frames included in the current situation data, the higher the resolution in the map server 2, but the greater the communication volume. The number of image frames included in the current situation data may be selected so that the data volume is below a predetermined upper limit. Alternatively, the system may be configured to extract only image areas where parked vehicles are reflected, rather than the entire image frame, and include them in the current situation data.

[0126] If step S205 is completed, step S206 is executed. In step S206, a data set including the vehicle behavior data acquired in step S203 and the current status data generated in step S205 is generated as a roadside parking location report and uploaded to the map server 2.

[0127] According to the above-mentioned structure, not only the behavior of the vehicle but also the recognition results and image data of the surrounding monitoring sensors can be collected on the map server 2. As a result, it is possible to verify with greater accuracy whether the parked vehicle on the roadside still exists or disappears. In addition, although the vehicle traveling in the second lane, that is, the vehicle traveling in the second lane does not take evasive action against the parked vehicle on the roadside, the parked vehicle may also be observed by the vehicle's front camera 11 and millimeter-wave radar 12. In other words, the parked vehicle on the roadside may also be observed by the vehicle traveling in the second lane. Although the probability and accuracy are reduced in the vehicles traveling in the third and fourth lanes, the parked vehicle may also be observed in the same way. According to the above-mentioned structure, the map server 2 can collect the sensing information collected from the vehicles traveling to the side of the parked vehicle on the roadside, so it can verify with greater accuracy whether the parked vehicle on the roadside exists.

[0128] In addition, while the above disclosure describes a method for uploading a roadside location report when the vehicle is traveling near a roadside location on a map, the upload process is not limited to this method. The map cooperation device 50 may also be configured to upload a roadside location report even when there is no roadside location on the map, for example, when vehicle behavior or sensing information indicating the presence of a parked vehicle is obtained.

[0129] like Figure 9 As shown, for example, the map cooperation device 50 may be configured to execute the process including steps S301 to S304. For example, the process may be executed independently of the upload process at a predetermined execution interval. Figure 9 In addition, for example, the process can also be executed when it is determined that there is no roadside parking location on the map during the upload process (No in step S102 or step S202). Figure 9 The processing flow shown.

[0130] In step S301, the vehicle behavior acquisition unit F3 acquires the vehicle's behavior for the most recent predetermined period (e.g., 10 seconds) from the vehicle state sensor 13 and the front camera 11, and then executes step S302. In step S302, it is determined whether the vehicle was traveling in the first lane during the aforementioned time period. If the vehicle was not traveling in the first lane, a negative determination is made in step S302, and the process ends. On the other hand, if the vehicle was traveling in the first lane, a positive determination is made in step S302, and the process proceeds to step S303.

[0131] In step S303, the time series data of the vehicle behavior data obtained in step S301 is analyzed to determine whether an avoidance action has been taken. For example, in the case of a change in the driving position accompanied by deceleration or stopping, or a sudden steering operation, it is determined that an avoidance action has been taken. It is also possible to determine whether the driving position has changed based on the trajectory of the vehicle's position. This can be determined based on the change in yaw rate, steering angle, lateral acceleration relative to time, the lighting status of the direction indicator, etc. In addition, it is also possible to determine whether the driving position has changed based on whether the lane dividing line has been crossed. Furthermore, it is also possible to determine that an avoidance action has been taken based on the yaw rate, steering angle, or lateral acceleration being above a specified value.

[0132] If it is determined in step S302 that an evasive action has been taken, the process proceeds to step S303, where a roadside parking location report is generated and transmitted, similar to steps S103 and S206 described above. Since parked vehicles are generally located on the roadside shoulder or in the first lane, vehicles traveling in the first lane are directly affected by the parked vehicle. The execution of an evasive action while traveling in the first lane indicates the presence of a parked vehicle. Furthermore, the roadside parking location report transmitted in step S303 corresponds to data indicating the presence of a parked vehicle that has not yet been identified by the map server 2.

[0133] The reporting location information for the roadside parking location report generated in step S303 may be set to the location of the vehicle immediately preceding the location where the avoidance action was taken. By setting the location of the vehicle preceding the location where the avoidance action was taken, the risk of incorrectly identifying a lane as containing a roadside parked vehicle can be reduced. Alternatively, a location a predetermined distance (e.g., 20 meters) in the direction of travel from the location of the vehicle preceding the location where the avoidance action was taken may be set as the reporting location.

[0134] In addition, if Figure 10 As shown, the map cooperation device 50 may also be configured to execute the processing including steps S401 to S403. Figure 10The processing flow shown may be executed independently of the upload process at a predetermined execution interval, or may be executed when it is determined during the upload process that there is no roadside stop on the map (No in step S102 or step S202).

[0135] In step S401, sensing information from the most recent specified time period (e.g., 5 seconds) is acquired, and step S402 is executed. In step S402, the roadside parking presence determination unit F51 analyzes the sensing information acquired in step S401 and the vehicle's behavior history to determine whether a roadside vehicle exists. If a roadside vehicle is determined to exist, a roadside parking location report is generated and uploaded, similar to step S206. Furthermore, the sensing information included in the roadside parking location report uploaded in step S403 can include, for example, recognition results and image frames from various surrounding monitoring sensors at the time the roadside vehicle was determined to exist. The roadside parking location report transmitted in step S403 is similar to the roadside parking location report transmitted in step S303 and serves as data notifying the map server 2 of the presence of a roadside vehicle that has not yet been identified.

[0136] <Map Server 2 Configuration>

[0137] Next, the configuration of map server 2 will be described. Map server 2 detects the presence and absence of parked vehicles on the roadside based on roadside parking location reports sent from multiple vehicles, and distributes this information to the vehicles as roadside parking location information. Map server 2 functions as a parking location management device. Furthermore, the description of the vehicle with which map server 2 communicates can be replaced with the in-vehicle system 1 or the map collaboration device 50.

[0138] like Figure 11 As shown, the map server 2 includes a server processor 21, a RAM 22, a storage device 23, a communication device 24, a map DB 25, and a vehicle position DB 26. The DB in the component name refers to a database. The server processor 21 is hardware for calculation and processing combined with the RAM 52. The server processor 21 is a structure including at least one CPU (Central Processing Unit) and other calculation cores. The server processor 21 performs various processes such as determining the existence status of roadside parked vehicles by accessing the RAM 22. The storage device 23 is a structure including a non-volatile storage medium such as a flash memory. The program executed by the server processor 21, namely the roadside parking information management program, is stored in the storage device 23. The server processor 21 executing the roadside parking information generation program is equivalent to executing the parking location management method corresponding to the roadside parking information management program. The communication device 24 is a device for communicating with other devices such as each vehicle-mounted system 1 via the wide area communication network 3.

[0139] The map database 25 is, for example, a database that stores high-precision map data. Furthermore, the map database 25 includes a roadside parking location database 251 that stores information related to locations where vehicles have been detected parked on the roadside. The map database 25 and the roadside parking location database 251 are implemented using rewritable non-volatile storage media. The server processor 21 is configured to allow data to be written, read, and deleted.

[0140] The roadside parking location DB251 stores data indicating locations where roadside parked vehicles are detected (hereinafter referred to as roadside parking location data). The roadside parking location data indicates the position coordinates of each roadside parking location, the lane where the roadside parked vehicle exists, the type of the roadside parked vehicle, the size, the amount of protrusion to the first lane, the time of appearance, the latest duration determination time, etc. The data on a certain roadside parking location is updated regularly by the roadside parking information management unit G3 based on the roadside parking location report from the vehicle for the location. The data of each roadside parking location constituting the roadside parking location data can be maintained in an arbitrary data structure such as a list form. For example, the data of each roadside parking location can also be stored separately according to the prescribed division. The division unit can be a grid of a high-precision map, an administrative division unit, or another division unit. For example, it can also be a road link unit. The grid of the map refers to a plurality of small areas formed by dividing the map according to a constant rule. The grid can also be replaced by a map tile.

[0141] The vehicle location DB26 is a database implemented using a rewritable non-volatile storage medium. The vehicle location DB26 is configured to be able to implement writing, reading, and deletion of data based on the server processor 21. In the vehicle location DB26, a correspondence is established with the vehicle ID to store data representing the current status of each vehicle including the location of the roadside parking information distribution system 100 (hereinafter referred to as vehicle location data). The vehicle location data represents the position coordinates, driving lane, direction of travel, driving speed, etc. of each vehicle. Whenever a vehicle status report is received from the vehicle, the data about a certain vehicle is updated by the vehicle location management unit G2 described later. The data of each vehicle constituting the vehicle location data can be maintained in an arbitrary data structure such as a list form. For example, the data of each vehicle can also be stored separately according to the specified division. The division unit can be a grid of a map, an administrative division unit, or another division unit (such as a road link unit).

[0142] The map server 2 provides the roadside parking information generation program stored in the storage device 23 by executing the roadside parking information generation program stored in the storage device 23 through the server processor 21. Figure 12The various functional modules shown in FIG. 2 correspond to functions. Specifically, the map server 2 includes a report data acquisition unit G1, a vehicle position management unit G2, a roadside parking information management unit G3, and a distribution processing unit G4. The roadside parking information management unit G3 includes a presence determination unit G31 and a disappearance determination unit G32.

[0143] The report data acquisition unit G1 acquires the vehicle status report and roadside parking location report uploaded from the in-vehicle system 1 via the communication device 24. The report data acquisition unit G1 provides the vehicle status report acquired from the communication device 24 to the vehicle position management unit G2. Furthermore, the report data acquisition unit G1 provides the roadside parking location report acquired from the communication device 24 to the roadside parking information management unit G3. The report data acquisition unit G1 functions as a judgment material acquisition unit.

[0144] The vehicle position management unit G2 updates the position information of each vehicle stored in the vehicle position DB 26 based on the vehicle status reports transmitted from each vehicle. Specifically, each time the report data acquisition unit G1 receives a vehicle status report, it updates the specified management items stored in the vehicle position DB 26, such as the location information of the source of the vehicle status report, the driving lane, the travel direction, and the driving speed.

[0145] The roadside parking information management unit G3 updates the data for each roadside parking location stored in the roadside parking location DB 251 based on roadside parking location reports transmitted from each vehicle. The appearance determination unit G31 and disappearance determination unit G32 within the roadside parking information management unit G3 are both components used to update the data for each roadside parking location. The appearance determination unit G31 is responsible for detecting the appearance of a roadside parked vehicle. The disappearance determination unit G32 is responsible for determining whether a roadside parked vehicle detected by the appearance determination unit G31 still exists, that is, whether the detected roadside parked vehicle has disappeared. The disappearance determination unit G32 determines the presence of a roadside parked vehicle at a parking location based on vehicle behavior data and sensor information received after the parking location is registered. Details of the appearance determination unit G31 and disappearance determination unit G32 will be described later. The appearance determination unit G31 corresponds to the parking location detection unit, and the disappearance determination unit G32 corresponds to the presence determination unit.

[0146] The distribution processing unit G4 is a component that distributes roadside parking location information. For example, the distribution processing unit G4 implements roadside parking location notification processing. This processing distributes a communication packet indicating the location of a parked vehicle (in other words, a roadside parking location), known as a roadside parking location notification packet, to vehicles scheduled to pass through the roadside parking location. The roadside parking location notification packet indicates the location coordinates of the parked vehicle and its characteristics (type, color), etc. The roadside parking location notification packet can be sent to vehicles scheduled to pass through the roadside parking location within a specified timeframe, such as one, two, or five minutes. For example, the planned travel path of each vehicle can be obtained to determine whether the vehicle is scheduled to travel through the roadside parking location. Alternatively, vehicles traveling on the same road / lane as, or connected to, the road / lane where the parked vehicle is located can be selected as vehicles scheduled to pass through the roadside parking location. The time required to reach the roadside stop can be calculated based on the distance from the current position of the vehicle to the roadside stop and the vehicle's traveling speed.

[0147] The delivery processing unit G4 uses road link and altitude information to select the destination for the roadside parking location notification packet. This reduces the risk of inadvertent delivery to vehicles traveling on roads running parallel to or above or below a road with parked vehicles. In other words, it can prevent erroneous identification of delivery targets on elevated roads or roads with double-deck structures. Delivery targets are extracted based on the location information and travel speed of each vehicle registered in the vehicle location DB 26.

[0148] Furthermore, by adding the time requirement until arrival at a roadside parking location to the distribution target extraction criteria, unnecessary distribution can be suppressed. This is because the presence of parked vehicles can change dynamically. Therefore, even if a distribution is made to a vehicle with more than 30 minutes until arrival, the parked vehicle is likely to have disappeared by the time it arrives. Furthermore, the time requirement until arrival at a roadside parking location is an optional element and need not be included in the distribution target extraction criteria.

[0149] Distribution targets can also be determined on a lane-by-lane basis. For example, vehicles traveling in the first lane are affected by parked vehicles. Vehicles currently traveling / scheduled to travel in the first lane are included in the distribution targets. On the other hand, vehicles traveling in the third or fourth lanes, which are farther away from the first lane, are less likely to be affected by parked vehicles. Therefore, vehicles scheduled to travel in lanes not adjacent to the first lane, such as the third lane, can be excluded from the distribution targets. Furthermore, vehicles traveling in the second lane, which need to be alert to vehicles entering from the first lane, can also be included in the distribution targets.

[0150] The roadside parking location notification packet can be distributed, for example, via multicast to multiple vehicles that meet the aforementioned distribution criteria. Alternatively, the roadside parking location notification packet can be distributed via unicast. When unicasting the roadside parking location notification packet, priority is given to vehicles closest to the roadside parking location, or vehicles arriving earlier, taking into account vehicle speed. Vehicles that are so close that they would not have time to respond to control or report the location of the roadside parking location, even if notified, may be excluded from the distribution targets.

[0151] Alternatively, the distribution processing unit G4 can be configured to transmit roadside parking location notification packets via a roadside device. In such a configuration, the roadside device broadcasts the roadside parking location notification packets received from the distribution processing unit G4 via narrow-area communication to vehicles within the communication area of ​​the roadside device. Alternatively, a regional multicast method can be used to distribute roadside parking location notification packets to vehicles within a specified distance of a registered obstacle location. Various methods can be used to distribute information.

[0152] Furthermore, the distribution processing unit G4 implements a disappearance notification process. Disappearance notification processing is the process of distributing a communication data packet (hereinafter referred to as a disappearance notification packet) indicating the disappearance of a roadside parked vehicle. For example, the disappearance notification packet can be distributed, for example, in a multicast manner, to vehicles that have received a roadside parking location notification packet. After the disappearance determination unit G32 determines that the roadside parked vehicle has disappeared, the disappearance notification packet is distributed as quickly as possible. Alternatively, the disappearance notification packet can be distributed in a unicast manner, similar to the roadside parking location notification packet. When distributing the disappearance notification packet in a unicast manner, the disappearance notification packets can be preferentially sent sequentially, starting with vehicles that are close to the roadside parking location or those that arrive earlier due to vehicle speed. Vehicles that are so close that they would not have time to react to or report the disappearance of the roadside parked vehicle can also be excluded from the distribution targets. Furthermore, since the distribution targets of the disappearance notification packet are limited to vehicles that have been notified of the existence of the roadside parked vehicle, the distribution targets are selected using road link and altitude information.

[0153] The distribution processing unit G4 may also manage information about vehicles to which roadside parking notification packets have been sent in the roadside parking location DB 251. By managing vehicles to which roadside parking notification packets have been sent, it is also possible to easily select recipients for the disappearance notification packet. Similarly, the distribution processing unit G4 may also manage information about vehicles to which disappearance notification packets have been sent in the roadside parking location DB 251. By managing whether a roadside parking location notification packet / disappearance notification packet has been received in the map server 2, it is possible to prevent repeated distribution of the same information. Alternatively, a flag or the like may be used on the vehicle side to manage whether a roadside parking location notification packet / disappearance notification packet has been received. Roadside parking location notification packets and disappearance notification packets correspond to roadside parking location information.

[0154] Server-side processing

[0155] use Figure 13 The flowchart shown in the figure explains the roadside parking location registration process performed by the map server 2. Figure 13 The preferred update cycle is a relatively short time such as five minutes or ten minutes.

[0156] In the map server 2, the server processor 21 repeatedly receives roadside stop location reports transmitted from vehicles at a constant frequency (step S501). Step S501 corresponds to the judgment material acquisition step. Upon receiving a roadside stop location report, the server processor 21 identifies the location for which the received roadside stop location report is a report (step S502) and stores the received roadside stop location reports by location (step S503). Furthermore, to account for variations in the location information reported in the roadside stop location reports, the roadside stop location reports may be stored in intervals of a specified length.

[0157] The server processor 21 then extracts locations that meet the specified update conditions (step S504). For example, locations where the number of reports received within a specified time period is above a specified threshold, and a specified waiting time has passed since the last determination of the presence / absence of a roadside parked vehicle, are extracted as update targets. The waiting time can be a relatively short time, such as three or five minutes. Alternatively, the update condition can be a location where the number of reports received is above a specified threshold, or a location where a specified waiting time has passed since the last update.

[0158] The conditions for executing the appearance determination process, described later, may differ from those for executing the disappearance determination process. The number of report receptions for executing the appearance determination process may also be less than the number of report receptions for executing the disappearance determination process. For example, the number of report receptions for executing the appearance determination process may be three, while the number of report receptions for executing the disappearance determination process may be six, twice as many. This configuration allows for rapid detection of the presence of parked vehicles and improves the accuracy of determining the disappearance of parked vehicles.

[0159] Once the update target locations have been extracted, any one of them is set as the target for processing (step S505), and it is determined whether it is a location registered as a roadside parking location or an unregistered location. If the target location is not registered as a roadside parking location, the appearance determination unit G31 performs appearance determination processing (step S507). Step S507 corresponds to the parking location detection step. On the other hand, if the target location is a location registered as a roadside parking location, the disappearance determination unit G32 performs disappearance determination processing (step S508). Step S508 corresponds to the existence determination step. Then, based on the determination result of the appearance determination processing or disappearance determination processing, the registration content of the roadside parking location DB 251 is updated (step S509).

[0160] For example, for locations where a parked vehicle is determined to be present, its information is added to the roadside parking location DB 251. For locations where a parked vehicle is determined to be absent, the location information is deleted from the roadside parking location DB 251, or a flag indicating disappearance is set. For roadside parking locations with a disappearance flag set, data can be deleted after a specified time (e.g., one hour) has passed since the flag was set. Furthermore, for locations whose existence status has not changed, changes to the registration content can be omitted. For locations whose existence status has not changed, only the time of determination can be updated to the latest information (i.e., the current time).

[0161] If the appearance determination process or disappearance determination process is completed for all the update target points extracted in step S504, this process ends. On the other hand, if there are any unprocessed points remaining, the unprocessed points are set as target points and the appearance determination process or disappearance determination process is executed (step S510).

[0162] <Occurrence determination processing>

[0163] The following describes the presence determination process performed by the presence determination unit G31. The presence determination unit G31 uses lane changes, acceleration and deceleration patterns of passing vehicles, camera images, detection results of parked vehicles by the onboard system 1, and patterns of traffic volume changes in each lane to determine whether a parked vehicle is present at a location to be determined. The term "location" herein encompasses the concept of an interval of a predetermined length.

[0164] For example, the presence determination unit G31 determines that a parked vehicle is present at a location where the number of lane changes from the first lane to the second lane that have been made within a certain period of time exceeds a predetermined threshold. The presence of lane changes can be determined using a determination result or report from the vehicle, or by detecting the presence of lane changes based on the vehicle's driving trajectory. Alternatively, the presence determination unit G31 can determine that a parked vehicle is present at a location where a predetermined number (e.g., three or more) of lane changes have been made consecutively.

[0165] For example, Figure 14 As shown, the position of the roadside parked vehicle based on the lane change can be determined based on the driving trajectory Tr1 with the latest lane change timing among the trajectories of multiple vehicles that have implemented lane changes. For example, it is determined that there is a roadside parked vehicle Pv at a location that is a specified distance (for example, 5m) further to the traveling direction side from the separation point (hereinafter referred to as the deepest separation point) Pd1 located closest to the traveling direction in the first lane. The separation point can be a location where the steering angle is above a specified threshold, or a location where the offset from the lane center is above a specified threshold. Or it can be a location where the lane dividing line begins to be crossed. The roadside parking location here has a specified width in the front-to-rear direction in order to allow a certain degree of error. The front-to-rear direction here is equivalent to the direction in which the road is extended.

[0166] In addition, the position of the roadside parked vehicle can also be determined based on the position of the recovery point Pe1 closest to the deepest separation point Pd1 (hereinafter referred to as the front recovery point). For example, it can also be the midpoint between the deepest separation point Pd1 and the front recovery point Pe1. The recovery point can be a location where the steering angle of a vehicle entering the first lane from the second lane using a lane change becomes less than a specified threshold. In addition, the recovery point can also be a location where the offset from the lane center of a vehicle entering the first lane using a lane change becomes less than a specified threshold. Instead of the steering angle, the angle of the vehicle body relative to the direction in which the road extends can be used. In addition, the position of the roadside parked vehicle can also be determined based on the detection position information included in the roadside parking location report. In the case where the detection position information of the roadside parked vehicle about the same roadside parking location is obtained from multiple vehicles, the average position of these detection position information can also be used as the position of the roadside parked vehicle.

[0167] However, as avoidance actions to avoid parked vehicles, it is common to implement a change of driving position from the first lane to the side of the second lane (hereinafter referred to as an avoidance position change) and a change of position to return to the center of the first lane (hereinafter referred to as a recovery position change) in groups. However, as shown in driving trajectory Tr1, a vehicle that changes lanes to the second lane to avoid a parked vehicle does not necessarily return to the first lane. For example, if a right turn is scheduled after passing the side of a parked vehicle, or if there is no free space to return to the original lane due to other vehicles, the vehicle does not return to the first lane. In addition, in manually driven vehicles, whether to return to the first lane changes depending on the mood and habits of the driver's seat passenger. In addition, as shown in driving trajectory Tr2, it is also possible that a vehicle traveling in the second lane changes lanes to the first lane after passing the side of a parked vehicle. The server processor 21 of the present disclosure does not count the number of vehicles that have performed both a breakaway position change and a recovery position change, but instead extracts locations where each type of position change is concentrated as roadside parking locations, thereby enabling more rapid detection of the presence of roadside parked vehicles. Of course, alternative methods may also be used to detect roadside parking locations based on the number of vehicles that have performed both a breakaway position change and a recovery position change.

[0168] In addition, if Figure 14 As shown, locations where parked vehicles exist on the roadside appear on the map as areas near the left and right road ends where no vehicle tracks are present (hereinafter referred to as trackless areas Sp). The presence determination unit G31 may also determine the presence of trackless areas Sp based on the track records of multiple vehicles within a specified time period. Furthermore, the presence determination unit G31 may also set locations that become trackless areas Sp as obstacle locations. In other words, the presence determination unit G31 may detect the presence of parked vehicles on the roadside based on the creation of trackless areas Sp.

[0169] Alternatively, the presence determination unit G31 may detect the presence of parked vehicles based on image data included in a roadside parking location report. For example, the presence of a parked vehicle may be determined based on the presence of parked vehicles in a lane confirmed by camera images from multiple vehicles. Alternatively, the presence determination unit G31 may determine the presence of a parked vehicle based on the detection results of parked vehicles by the roadside parking presence determination unit F51, as included in the roadside parking location report. For example, if the number of reports indicating the presence of parked vehicles within a recent predetermined period exceeds a predetermined threshold, the presence determination unit G31 may determine that a parked vehicle exists at the location from which the reports were sent. Furthermore, the presence determination unit G31 may determine the presence of a parked vehicle based on the detection results of parked vehicles by surrounding monitoring sensors, as included in the roadside parking location report. For example, based on the detection results from multiple vehicles, a parked vehicle that has been parked at the same location for more than five minutes may be determined as a parked vehicle.

[0170] In addition, the occurrence determination unit G31 can also detect a location in the first lane where a specified acceleration / deceleration pattern occurs as a roadside parking location. Typically, the driver's seat passenger / autonomous driving system, upon recognizing the presence of a roadside parked vehicle in front of the vehicle, temporarily decelerates, changes the driving position, and then accelerates. In other words, it is assumed that an acceleration / deceleration pattern of deceleration followed by acceleration is observed near the roadside parking location. Conversely, an area where the frequency / number of consecutive occurrences of the above-mentioned acceleration / deceleration pattern within a recent specified time period exceeds a specified threshold can also be extracted as a roadside parking location. The change in driving position here includes not only lane changes, but also actions that bring the driving position in the first lane closer to the side of the second lane, and driving across the dividing line between the first and second lanes.

[0171] Furthermore, even when a moving object such as a bird, pedestrian, or wild animal is momentarily parked on the roadside, an acceleration / deceleration pattern of temporary deceleration followed by acceleration may be observed. In such cases, it is preferable to use areas associated with changes in driving position as parent groups for detecting roadside parking locations using the acceleration / deceleration pattern. In other words, it is preferable for the presence determination unit G31 to detect areas where a predetermined acceleration / deceleration pattern is observed along with changes in driving position as roadside parking locations.

[0172] While the above disclosure utilizes forward and backward acceleration for detecting roadside parking locations, it is assumed that a predetermined pattern of lateral acceleration also occurs when the driving position is changed to avoid parked vehicles. For example, areas where the frequency or number of consecutive occurrences of a predetermined acceleration and deceleration pattern in the left and right directions within a recent predetermined period exceeds a predetermined threshold may be selected as roadside parking locations.

[0173] Alternatively, if a parked vehicle is present, it can be expected that the traffic volume in the first lane will decrease compared to the traffic volume in the second lane. Alternatively, if the traffic volume in the first lane decreases by a predetermined value / ratio within a recent predetermined period compared to a predetermined period, and the traffic volume in the second lane increases during the same period, it can be determined that a parked vehicle is present in the first lane. Furthermore, the location of the parked vehicle detected in the first lane using the above method can be determined based on the trajectory of vehicles traveling in that lane.

[0174] Furthermore, the presence determination unit G31 may also detect a parked vehicle location based on the autonomous driving device handing over authority to the driver's seat passenger or the driver's seat passenger overriding the vehicle. For example, the presence of a parked vehicle may be detected by acquiring and analyzing images from the front camera 11 when the autonomous driving device transfers authority to the passenger or detects an overriding driver's seat passenger, and determining whether the cause is a parked vehicle.

[0175] While multiple perspectives for determining the presence of a parked vehicle have been listed above, the presence determination unit G31 may use any of these perspectives to determine the presence of a parked vehicle. Furthermore, a combination of multiple perspectives may be used to determine the presence of a parked vehicle. When using a combination of multiple perspectives to determine the presence of a parked vehicle, weights corresponding to the type of judgment criteria may be assigned. For example, if the weight for avoidance action is 1, the recognition result from a single camera may be assigned a weight of 1.2, and the fusion recognition result may be assigned a weight of 1.5.

[0176] In addition, if Figure 15 As shown, the threshold value for the number of vehicles that have taken avoidance action for determining the presence of parked vehicles on the roadside can also be changed based on whether the presence of parked vehicles on the roadside is confirmed as a result of the analysis of the image provided by the server processor 21 from the vehicle. The avoidance action here refers to the action of moving the driving position from the first lane to the second lane side, and the number of vehicles that have taken avoidance action will be counted as vehicles that have traveled in the first lane until they are close to the roadside parking location in the parent group. The operator can also perform image-based confirmation of the presence of parked vehicles on the roadside. In addition, the number of vehicles that have taken avoidance action required for determining the presence of parked vehicles on the roadside can also be changed based on whether parked vehicles are detected in the vehicle's peripheral monitoring sensors or the roadside parking determination unit F51. In addition, Figure 15The "Number of Vehicles" column can be replaced with the ratio of vehicles that implemented evasive action or the number of consecutive roadside parking location reports indicating that evasive action was implemented. Alternatively, if image analysis confirms that a parked vehicle is completely confined to the roadside shoulder area outside the lane, even if there is no vehicle behavior data indicating evasive action, it may still be determined that a parked vehicle is present. However, in this case, it is preferably registered in the roadside parking location DB 251 in association with a protrusion amount into the lane of zero.

[0177] <Disappearance determination processing>

[0178] The disappearance determination process performed by the disappearance determination unit G32 will now be described. Based on roadside parking location reports, the disappearance determination unit G32 periodically determines whether a roadside vehicle still exists at a roadside parking location detected by the presence determination unit G31. Information used to determine whether a roadside vehicle has disappeared can include the presence or absence of lane changes, the vehicle's driving trajectory, the acceleration and deceleration patterns of passing vehicles, camera images, the results of the onboard system 1 identifying roadside vehicles, and the changing patterns of traffic volume in each lane.

[0179] For example, the disappearance determination unit G32 can make a determination based on a decrease in the number of lane changes executed at a roadside parking location. For example, the disappearance determination unit G32 can determine that the parked vehicle has disappeared when the number of lane changes near the roadside parking location falls below a predetermined threshold. Alternatively, the disappearance determination unit G32 can compare the decrease in the number of lane changes near the roadside parking location as a vehicle behavior with the time when the parked vehicle was detected, and determine that the parked vehicle has disappeared when a statistically significant difference is found.

[0180] The disappearance determination unit G32 may also determine that a parked vehicle has disappeared based on a decrease in the number of vehicles traveling across the boundary between the first and second lanes near the roadside parking location. Alternatively, the disappearance determination unit G32 may determine that a parked vehicle has disappeared based on the average value of the offset from the lane center of the first lane toward the second lane falling below a predetermined threshold. In other words, the disappearance determination unit G32 may determine that a parked vehicle has disappeared based on the change in lateral position of vehicles traveling near the roadside parking location falling below a predetermined threshold.

[0181] The disappearance determination unit G32 may also determine that a parked vehicle has disappeared based on the presence of a vehicle that passes through (i.e., drives straight ahead) the roadside stop without performing an evasive action such as a lane change. For example, the presence of a vehicle traveling through the roadside stop can be determined based on its driving trajectory. More specifically, the disappearance of a parked vehicle may be determined when the driving trajectory of a vehicle passes through the roadside stop. Alternatively, the disappearance of a parked vehicle may be determined when the number of parked vehicles exceeds a predetermined threshold.

[0182] Furthermore, when the roadside parking location report includes a camera image, the disappearance determination unit G32 may also analyze the camera image to determine whether the parked vehicle still exists. The disappearance determination unit G32 may also perform statistical processing on the analysis results of image data from multiple vehicles to determine whether the parked vehicle still exists. This statistical processing includes majority decision and averaging.

[0183] When the roadside parking location report includes the determination result of the roadside parking presence determination unit F51, the disappearance determination unit G32 may also perform statistical processing on the roadside parking presence determination results for multiple vehicles to determine whether the roadside parked vehicle still exists or has disappeared. For example, the roadside parked vehicle may be determined to have disappeared when the number of received reports indicating the absence of the roadside parked vehicle exceeds a predetermined threshold. Furthermore, when the roadside parking location report includes the detection results of parked vehicles by the surrounding monitoring sensors, the disappearance determination unit G32 may also perform statistical processing on the detection results of parked vehicles for multiple vehicles to determine whether the roadside parked vehicle still exists or has disappeared.

[0184] The disappearance determination unit G32 may also determine that the parked vehicle has disappeared if no predetermined acceleration / deceleration pattern is observed as the behavior of a vehicle passing near the roadside stop. Alternatively, the parked vehicle may be determined to have disappeared if the traffic volume between the first lane and the second lane has ceased to have a significant difference, or if the difference has decreased, resulting in an increase in the traffic volume in the first lane. Traffic volume can be, for example, the number of vehicles passing per unit time within a road section from the roadside stop to 400 meters immediately preceding the stop.

[0185] While multiple perspectives for determining whether a parked vehicle has disappeared have been listed above, the disappearance determination unit G32 may use any one of the aforementioned perspectives to determine whether a parked vehicle has disappeared, or may use a combination of multiple perspectives to determine whether a parked vehicle has disappeared. When using a combination of multiple perspectives to determine whether a parked vehicle has disappeared, a weight corresponding to the type of judgment criteria may be added to the determination.

[0186] In addition, if Figure 16As shown, the threshold value for determining the disappearance of a parked vehicle on the roadside, such as the number of vehicles going straight at the target location, can also be changed based on whether the disappearance of the parked vehicle is confirmed as a result of the analysis of the image provided by the server processor 21 from the vehicle. In addition, the image analysis (image confirmation) can be performed by an operator instead of the server processor 21. In addition, the threshold value for the number of vehicles going straight at the location required for determining the disappearance of a parked vehicle on the roadside can also be changed based on whether a parked vehicle is detected in the vehicle's peripheral monitoring sensor or the roadside parking determination unit F51. Going straight here means traveling along the road in the first lane without changing the driving position from the first lane to the second lane side. Going straight here does not necessarily mean driving while maintaining the steering angle at 0°. In addition, for the number of vehicles going straight, the vehicles traveling in the first lane can be counted in the parent group. In addition, Figure 16 The column of the number of vehicles in can be replaced with the ratio of vehicles that go straight into the location or the number of times that roadside parking location reports indicating that there are no parked vehicles on the roadside are received consecutively.

[0187] Furthermore, if, despite the presence of parked vehicles as a result of image verification of the roadside parking location by the server processor 21, many vehicles are driving straight into the location, the roadside parking information management unit G3 may determine that the parked vehicles are confined to the roadside strip and do not protrude into the first lane. The roadside strip here refers to the area outside the lane. The roadside strip may also include the concept of the road shoulder.

[0188] <Supplementary information on the method for determining the appearance and disappearance of parked vehicles on the roadside>

[0189] Static map elements such as road structures rarely change over time, so map data updates for these elements can utilize numerous driving trajectories accumulated over a specified period, such as a week or a month. By updating map data using reports from numerous vehicles as a central cluster, improved accuracy can be expected.

[0190] However, parked vehicles represent dynamic map elements whose presence and absence changes relatively quickly compared to, for example, road structures. Therefore, detecting the presence and absence of parked vehicles requires greater real-time performance. To improve the accuracy of information such as the presence and location of parked vehicles, it is preferable to use reports from many vehicles as a parent group. However, collecting reports from a larger number of vehicles takes time, compromising real-time performance. In other words, to ensure real-time detection of the presence and absence of parked vehicles compared to static map generation, it is necessary to determine and distribute them with the highest possible accuracy based on fewer vehicle reports.

[0191] In this context, the presence determination unit G31 detects roadside parking locations based on, for example, roadside parking location reports received within a predetermined first time period from the current time. Furthermore, the disappearance determination unit G32 determines the presence / disappearance of parked vehicles based on roadside parking location reports received within a predetermined second time period. To ensure real-time performance, both the first and second time periods are preferably set to be shorter than ninety minutes, for example. For example, the first time period can be set to ten, twenty, or thirty minutes. The second time period can also be ten, twenty, or thirty minutes. The first and second time periods can be the same or different in length. The first and second time periods can also be five minutes or one hour, for example.

[0192] Information about the presence of parked vehicles is considered more useful for driving control than information about their disappearance. This is because if information about lanes with parked vehicles can be acquired in advance as map data, sufficient avoidance maneuvers can be planned and implemented. Consequently, there is also a need to detect and notify the presence of parked vehicles more quickly. Based on this, the first time period can be set shorter than the second time period to enable earlier detection and notification of the presence of parked vehicles.

[0193] Furthermore, it is also assumed that there is a need to avoid erroneous determination that a parked vehicle has disappeared, even though it still exists, and thus causing erroneous distribution. To address this need, the second time period may be set longer than the first time period. By setting the second time period longer than the first time period, the presence of a parked vehicle can be promptly notified, and the risk of erroneous determination that a parked vehicle has disappeared can be reduced.

[0194] The presence determination unit G31 and the disappearance determination unit G32 can also be configured to prioritize information indicated by more recent reports, for example by increasing weights, to determine the presence / presence of parked vehicles. For example, information acquired within ten minutes can be weighted 1, information acquired within thirty minutes but more than ten minutes ago can be weighted 0.5, and information acquired further in the past can be weighted 0.25. Statistical processing can then be performed by applying weighting coefficients corresponding to the recency of the information. This configuration allows the most recent status to be more strongly reflected in the determination results, improving real-time performance.

[0195] In addition, the roadside parking information management unit G3 can also perform weighting according to the characteristics of the reporting source and statistically process the reports from multiple vehicles to determine the presence and absence of roadside parked vehicles. For example, the weight of the reports from the autonomous driving vehicle can be set to be larger than the reports from the manually driven vehicle. It can be expected that the autonomous driving vehicle is relatively equipped with high-performance millimeter wave radar 12, front camera 11, LiDAR, etc. In addition, the possibility of the autonomous driving vehicle unnecessarily changing the driving position is low. The change in the driving position of the autonomous driving vehicle is relatively likely to be an action to avoid roadside parked vehicles. Therefore, by giving priority to the use of reports from the autonomous driving vehicle, the accuracy of determining the presence or absence of roadside parked vehicles can be improved.

[0196] Furthermore, the appearance determination unit G31 and the disappearance determination unit G32 may be configured to treat reports from vehicles that frequently change their driving position, such as lane changes, as noise and not use them in the determination process. The vehicle position management unit G2 may identify unstable vehicles based on sequentially uploaded vehicle status reports and manage such vehicles using signs, etc. This configuration reduces the risk of misjudging the presence of parked vehicles based on reports from vehicles driven by users who frequently change lanes. A variety of conditions can be applied to the criteria for determining whether a vehicle is considered to have an unstable driving position. For example, a vehicle whose number of lane changes within a certain period of time exceeds a specified threshold can be identified as an unstable driving position vehicle. This threshold is preferably set to three or more in order to exclude lane changes (two lane changes, one for disengagement and one for recovery) used to avoid parked vehicles. For example, an unstable driving position vehicle may be a vehicle that changes lanes four or more times within a certain period of time, such as ten minutes.

[0197] In addition, if Figure 15 as well as Figure 16 As illustrated, the conditions (e.g., thresholds) for determining the presence of a parked vehicle may be different from the conditions for determining the disappearance of a parked vehicle. For example, the conditions for determining the disappearance of a parked vehicle may be set more strictly than the conditions for determining the presence of a parked vehicle. The judgment material for the presence of a parked vehicle may be different from the judgment material for the disappearance of a parked vehicle. In addition, the weight of each type of information may be different when determining the presence and when determining the disappearance. For example, when determining the presence of a parked vehicle, the weight of the analysis result of the camera image may be greater than the vehicle behavior data, while on the other hand, when determining the disappearance of a parked vehicle, the weight of the vehicle behavior data may be greater than the analysis result of the camera image. This is because the camera image is intended for verification of the presence of an object, while on the other hand, if the possibility of shooting other places is considered, the reliability of verification without an object is poor.

[0198] Vehicle Control Processing

[0199] Next use Figure 17 An example of vehicle control using roadside parking information is described. For example, it is executed independently of the above-mentioned upload process, etc. Figure 17 For example, when the automatic lane change function of the driving assistance ECU 60 is enabled based on a user operation, the automatic lane change function is executed at a predetermined period. Figure 17 The vehicle control process shown in the figure is sufficient. In addition, the state in which the automatic lane change function is enabled also includes automatic driving in which the vehicle autonomously drives according to a predetermined driving plan. Figure 17 The vehicle control process shown includes steps S601 to S608 as an example. The driving support ECU 60 and the map cooperation device 50 cooperate to execute steps S601 to S608.

[0200] First, in step S601, the map cooperation device 50 reads the roadside parking information on the map stored in memory M1, provides it to the driving assistance ECU 60, and then moves on to step S602. In step S602, the driving assistance ECU 60 determines whether there is a roadside parking vehicle within a specified distance ahead of the vehicle's driving lane based on the roadside parking information on the map. If there is no roadside parking vehicle, step S602 is denied and the process ends. In this case, driving control continues based on the separately prepared driving plan. On the other hand, if there is a roadside parking vehicle, step S602 is affirmatively determined and step S603 is executed.

[0201] In step S603, a determination is made as to whether the vehicle is traveling in the first lane based on the output signal from the front camera 11. If the vehicle is traveling in the first lane, a positive determination is made in step S603, and the process proceeds to step S604. On the other hand, if the vehicle is not traveling in the first lane, a negative determination is made in step S603, and the process proceeds to step S607.

[0202] In step S604, the driving plan is revised to become a driving trajectory that avoids parked vehicles on the roadside. For example, a driving plan is created that includes a lane change from the first lane to the second lane. The revised driving plan also includes the setting of the point of departure from the first lane to the second lane (that is, the lane change point). If step S604 is completed, step S605 is executed. In step S605, the HMI system 16 is cooperated with to prompt information related to the revised driving plan. For example, the passenger is notified of the implementation of the lane change to avoid parked vehicles on the roadside. If step S605 is completed, the process moves to step S606. In step S606, the lane change is executed and the present process ends.

[0203] In step S607, based on the output signal of the front camera 11, it is determined whether the vehicle is traveling in the second lane. If traveling in the second lane, step S607 is determined to be positive and step S608 is executed. On the other hand, if not traveling in the second lane, step S607 is determined to be negative and the current process ends. In step S608, an interruption warning process is executed. For example, as an interruption warning process, in cooperation with the HMI system 16, a message urging attention to interruptions from the first lane is displayed. In addition, as an interruption warning process, the inter-vehicle distance to the preceding vehicle can also be set longer to make it easier for the vehicle traveling in the first lane to enter the second lane.

[0204] Furthermore, while the above disclosure describes a configuration in which no special processing is performed when the vehicle's driving lane is not in either the first or second lane, the present invention is not limited to this. Notification of the presence of parked vehicles may also be provided even when the vehicle is in the third lane. Furthermore, preferably, the further the driving lane is from the first lane, the weaker the information provided regarding parked vehicles. This is because the further the driving lane is from the first lane, the less likely it is to be affected by parked vehicles, and the less useful the information is to the user. By weakening the information provided regarding parked vehicles as the driving lane is further from the first lane, the user's concern about the risk of being annoyed can be reduced.

[0205] An example of the system's operation and effects

[0206] According to the above system configuration, the map collaboration device 50 first triggers the upload of a roadside parking location report using the detection of a parked vehicle. Based on the information uploaded from the vehicle, the map server 2 detects the location of the parked vehicle (i.e., the roadside parking location). It then notifies vehicles scheduled to travel near the roadside parking location of the parked vehicle. Furthermore, the map collaboration device 50 transmits to the map server 2 at least one of vehicle behavior data representing the vehicle's behavior while near the roadside parking location notified by the map server 2, and sensing information.

[0207] Here, assuming that a parked vehicle remains and the vehicle is traveling in a lane such as the first lane that is affected by the parked vehicle, the vehicle behavior data sent by the map cooperation device 50 to the map server 2 indicates that an avoidance action has been taken. In addition, even if the vehicle is traveling in a lane without a parked vehicle, it may be possible to decelerate in order to avoid a collision with a vehicle that has changed lanes to avoid the parked vehicle. In other words, deceleration and re-acceleration to avoid a collision with an intervening vehicle may be observed, which are behaviors that cannot be observed during cruising. On the other hand, when the parked vehicle disappears, the vehicle behavior to avoid the parked vehicle or the intervening vehicle is not observed. In other words, the vehicle behavior data near the registration location of the parked vehicle acts as an indicator of whether the parked vehicle remains.

[0208] Furthermore, if it is assumed that a parked vehicle remains, sensing information indicating the presence of the parked vehicle or including an image of the parked vehicle is transmitted to the map server 2. On the other hand, if the parked vehicle has disappeared, the sensing information is not included in the sensing information. Thus, the sensing information also serves as an indicator of whether a parked vehicle remains. In other words, vehicle behavior data near the roadside parking location and sensing information serve as indicators of whether a parked vehicle remains.

[0209] Therefore, the map server 2 can determine whether a parked vehicle remains at a roadside parking location or has disappeared based on at least one of vehicle behavior data and sensor information provided by multiple vehicles. Furthermore, if the disappearance of a parked vehicle is detected based on reports from vehicles passing through the roadside parking location, the vehicle to which the roadside parking information has been distributed can be distributed. This configuration enables the real-time identification of the location and continued existence of parked vehicles. Furthermore, the map server 2 can also detect the movement of a parked vehicle if it fails to receive roadside parking location reports from vehicles.

[0210] Figure 18 This is a diagram schematically showing the change in vehicle behavior caused by the presence or absence of roadside parking location information on the map. Figure 18 As shown in (A), the front camera 11 is able to identify the position of the parked vehicle on the roadside, and then implements avoidance actions such as lane changes. The map server 2 collects such vehicle behavior, checks the existence / appearance of the parked vehicle on the roadside, and starts distributing it as roadside parking location information. The recognizable position may change depending on the performance of the front camera 11, the millimeter wave radar 12, and the size of the parked vehicle on the roadside. Under good conditions such as sunny days, the recognizable position is a location close to the parked vehicle, for example, about 100m to 200m.

[0211] Figure 18 (B) schematically shows the behavior of a vehicle that has obtained roadside parking location information from the map server 2. Figure 18 As shown in (B), a vehicle that has acquired roadside parking location information as map data from the map server 2 can start lane changes before reaching a recognizable position. In other words, lane changes, handovers, and other responses can be implemented with ease.

[0212] On the other hand, the parked vehicle disappears as time passes by, such as by moving. There is a predetermined time difference (i.e., delay) between the disappearance of the parked vehicle in the real world and the detection of the situation by the map server 2. Therefore, if the parked vehicle disappears immediately after the real world, Figure 18 As shown in (C), a scenario occurs in which a vehicle passes by after changing lanes based on the roadside parking information on the map, even though there is actually no roadside parked vehicle.

[0213] However, the map server 2 of the present disclosure is configured to receive roadside parking location reports from vehicles passing near roadside parking locations. Therefore, the absence of parked vehicles can be quickly identified based on these roadside parking location reports. Consequently, the absence of parked vehicles can be quickly communicated to vehicles, reducing the risk of vehicles making unnecessary lane changes or handovers. Figure 18 (D) shows the state after the map server 2 confirms the disappearance of the parked vehicle on the roadside.

[0214] In addition, the map server 2 of the present disclosure verifies whether the parked vehicle has truly disappeared based on reports from multiple vehicles and / or from multiple perspectives. This configuration can reduce the worry that a parked vehicle may be mistakenly reported as disappeared despite its actual presence.

[0215] Furthermore, according to the configuration of the present disclosure, when a determination is made that a parked vehicle has disappeared as a result of analyzing an image uploaded from a vehicle, the threshold for the number of vehicles that do not take evasive action based on the determination that the parked vehicle has disappeared is lowered. Furthermore, when a determination is made that a parked vehicle has disappeared by statistically processing the sensing information from the perimeter monitoring sensors of multiple vehicles, the threshold for the number of vehicles that do not take evasive action based on the determination that the parked vehicle has disappeared is lowered.

[0216] For example, when the image analysis result in the server processor 21 confirms that the parked vehicle has disappeared, it is also possible to determine that the parked vehicle has disappeared based on the behavior information of one or more vehicles. According to this configuration, the determination of the disappearance of the parked vehicle can be confirmed more quickly. As a result, for example, the time from Figure 18 The transition period from (C) to (D) is achieved by combining vehicle behavior and image analysis to determine the presence status of parked vehicles on the roadside, achieving both real-time and information reliability.

[0217] Furthermore, as an example, the map server 2 may determine that a parked vehicle has disappeared based on the condition that the vehicle does not take any action to avoid the parked vehicle. This configuration eliminates the need for image-based determination, thereby reducing the risk of erroneous determination that a parked vehicle has disappeared when the camera does not capture the parked vehicle.

[0218] Furthermore, the above configuration detects parked vehicles on ordinary roads (i.e., parked vehicles on the roadside). Parked vehicles on ordinary roads can sometimes block nearly half of the first lane, potentially hindering automated driving / driving assistance functions. For example, a parked vehicle blocking the first lane could disrupt automated driving services. By distributing the locations of parked vehicles as described above, it is possible to perform handovers with sufficient efficiency or adopt a route without parked vehicles.

[0219] Furthermore, unlike stationary objects such as fallen objects, parked vehicles on typical roads often experience relatively dynamic changes in their presence and presence, as they are caused by the user of the vehicle, which is different from stationary objects such as fallen objects. In other words, the detection and disappearance of parked vehicles requires greater real-time performance than that of fallen objects.

[0220] Furthermore, in urban areas, parked vehicles occur more frequently than fallen objects. In particular, parked vehicles are more likely to occur around stations and facilities. Furthermore, the area affected by parked vehicles expands and contracts over time, with new parked vehicles appearing before and after a parked vehicle. Furthermore, parked vehicles tend to occur more frequently during the day when people are more active, but less frequently during the late night and early morning hours. Alternatively, it is possible to assume that there are areas where parked vehicles are not present during the day due to legal restrictions, but where they are more likely to appear late at night when restrictions are lifted. In other words, parked vehicles have different characteristics as obstacles, depending on the time of day and location.

[0221] Furthermore, since parked vehicles are still vehicles, it is difficult to distinguish whether they are parked or temporarily parked, such as waiting for a signal, based solely on images. In contrast, a system for detecting parked vehicles using information indicating the relative position of a parked vehicle to road structures, such as the distance to the road end and the distance to a traffic light, can improve the accuracy of distinguishing between parked vehicles and vehicles waiting for a signal.

[0222] In addition, according to the above-mentioned configuration, the vehicle-mounted system 1 can also pre-check the presence of parked vehicles on the roadside in places that are likely to become blind spots for surrounding monitoring sensors, such as after turning right or left at an intersection, near the exit of a curve, etc., based on the distributed information from the map server 2. As a result, the automatic driving device or the driving assistance ECU 60 can make appropriate path planning. Specifically, assume that as a path planning for turning left at an intersection, an automatic driving device makes a path planning that enters the first lane after turning left as the basic path planning. In such an automatic driving device, if the first lane after turning left is blocked by parked vehicles on the roadside, it is necessary to change the path planning within the intersection. For example, a path is generated that enters the second lane instead of the first lane.

[0223] If information about parked vehicles is not available as map data, the autonomous driving device described above could detect a parked vehicle after exiting an intersection and abruptly change its route plan, potentially causing the user to experience unpleasant lateral acceleration. Alternatively, the driver could suddenly change driving position, potentially confusing drivers of surrounding vehicles. To address these issues, the autonomous driving device utilizing the roadside parking information distribution system disclosed herein can pre-plan a route based on the presence of parked vehicles after a left turn. This reduces the user's concern about experiencing lateral acceleration exceeding a specified value or confusing drivers of surrounding vehicles. While the above describes the effects of the autonomous driving device planning a route, the same applies to human-driven driving. By notifying the user of the presence of parked vehicles in blind spots such as after a right or left turn using the HMI system 16, driving operations based on these circumstances can be performed. In other words, the system disclosed herein improves the stability of vehicle control by the autonomous driving device and the driver, as well as user convenience.

[0224] In addition, according to the above configuration, the vehicle-mounted system 1 can obtain whether the shoulder of the road is blocked by parked vehicles on the roadside. The automatic driving device is equipped with a function to use MRM (Minimum Risk Maneuver) to retreat to a place where it can be safely parked, such as the shoulder of the road, when it is difficult to continue the automatic driving. There is also the possibility that a place where it should be safe to park on a general road is temporarily unavailable due to parked vehicles on the roadside. The parking location information distributed by the above-mentioned roadside parking information distribution system 100 can be used for destination setting during MRM. For example, when the vehicle-mounted system 1 is executing MRM and there are all parked vehicles in the nearest safe parking lot, it can take measures such as searching for other places or parking in the current driving lane with slow deceleration.

[0225] The embodiments of the present disclosure have been described above, but the present disclosure is not limited to the above-mentioned embodiments. The various supplementary matters and modified examples described below are also included in the technical scope of the present disclosure, and various changes can be made to the embodiments without departing from the main purpose except for the following. For example, the various structures described below can be appropriately combined and implemented within the scope that does not cause technical contradictions. In addition, the same figure marks are attached to the components having the same functions as the components described in the above-mentioned embodiments, and their descriptions are omitted. In addition, when only a part of the structure is mentioned, the structure of the previously described embodiment can be applied to the other parts.

[0226] <Additional information on the appearance / disappearance determination process for roadside parking areas>

[0227] When the map cooperation device 50 recognizes a plurality of vehicles parked in a row on the road based on camera images, it is difficult to determine whether these vehicles are a row of vehicles waiting for a signal, that is, a temporarily parked row, or a row of vehicles parked on the roadside. However, when the detected vehicle row includes vehicles whose body orientation is opposite to that of other vehicles, the possibility that the vehicle row is a temporarily parked row is extremely small. A vehicle row with inconsistent orientations can also be determined as not a row of vehicles waiting for a signal, but a row of parked vehicles. The orientation of the vehicle can be determined by image analysis. In this way, the roadside parked vehicle presence determination unit F51 can also determine whether each vehicle is equivalent to a roadside parked vehicle based on whether the body orientations of these vehicles are consistent when a row of parked vehicles is detected. In addition, the roadside parked vehicle presence determination unit F51 can also determine that a stopped vehicle stored in the rearview mirror is a roadside parked vehicle.

[0228] The roadside parking presence determination unit F51 may also determine the position of the stopped vehicle detected by the surrounding monitoring sensor on the map data as described above, thereby identifying whether the stopped vehicle is a roadside parked vehicle or a temporarily stopped vehicle waiting for a traffic light. More specifically, the roadside parking presence determination unit F51 may also determine whether the stopped vehicle is a roadside parked vehicle or a temporarily stopped vehicle waiting for a traffic light. Figure 19 In the sequence shown, it is determined whether the detected stopped vehicle is a vehicle temporarily stopped due to waiting for a traffic light, etc., or a vehicle parked on the roadside. Figure 19 The flowchart shown includes steps S701 to S709 as an example. The number of steps in the flowchart, the processing order, etc. can be changed as appropriate. This processing flow can be implemented in parallel, in combination, or in substitution with the various processes described above.

[0229] First, in step S701, the map acquisition unit F2 acquires map data of a predetermined range based on the current position from the map storage unit 143 disposed inside or outside the vehicle, and the process proceeds to step S702. The map data read here may at least include information indicating the position of a road end.

[0230] In step S702, the detected object information acquisition unit F4 acquires sensing information from the surrounding monitoring sensors and the process proceeds to step S703. In step S703, the vehicle position acquisition unit F1 performs positioning processing based on information such as landmarks included in the sensing information and map data, and the process proceeds to step S704. Alternatively, step S704 may simply be a process for acquiring positioning results from a GNSS receiver.

[0231] In step S704, the roadside parking presence determination unit F51 determines whether a stopped vehicle has been detected based on the sensing information acquired in step S702. If no stopped vehicle has been detected, step S704 is negatively determined, and the process ends. On the other hand, if a stopped vehicle has been detected, the process proceeds to step S705. In step S705, the detected stopped vehicle's relative position relative to the host vehicle and the host vehicle's position information determined in step S703 are used to determine the stopped vehicle's position on the map, and the process proceeds to step S706. This process enables the calculation of the distance from the road end to the stopped vehicle using map data.

[0232] In step S706, based on the calculation results from step S705, a determination is made as to whether the distance from the road end to the stopped vehicle is less than a predetermined parking determination threshold. The parking determination threshold here is a parameter used to distinguish whether the detected stopped vehicle is parked on the roadside or temporarily parked, such as at a traffic light, and corresponds to the second distance described above. The parking determination threshold can be set to, for example, 0.75m or 1m.

[0233] If the distance from the road end to the stopped vehicle is less than the parking determination threshold, the stopped vehicle is determined to be parked on the roadside in step S707, and the process proceeds to step S708. In step S708, a roadside parking location report containing information about the parked vehicle is generated and transmitted to the map server 2, ending the process.

[0234] On the other hand, if the distance from the road end to the stopped vehicle is greater than the parking determination threshold, in step S709 it is determined that the stopped vehicle is not a roadside parked vehicle but a temporarily stopped vehicle waiting for a traffic light, etc., and this process ends.

[0235] The above configuration is equivalent, in one aspect, to using map data to estimate whether a stopped vehicle is more than a specified distance from the road edge, and based on this estimation result, identifying whether the detected stopped vehicle corresponds to a vehicle parked at the roadside. Generally speaking, if there are only one or so stopped vehicles, or if there is ample space in front and behind the stopped vehicles, it is expected that the perimeter monitoring sensors will be able to detect the road edge. On the other hand, if the stopped vehicles are arranged in a row with relatively close spacing, it may be difficult for the perimeter monitoring sensors to detect road information outside the row of stopped vehicles. This is because the stopped vehicles may become obstructions, preventing the perimeter monitoring sensors from detecting the road edge. Therefore, it may be difficult to calculate the distance from the road edge to the stopped vehicle using only the perimeter monitoring sensors.

[0236] To address this issue, as described above, by utilizing static map data that includes location information such as road edges, it is possible to improve the accuracy of determining whether a detected stopped vehicle corresponds to a parked vehicle on the roadside, even in situations where the perimeter monitoring sensors cannot or have difficulty detecting the location of road edges. Furthermore, the term "outside the row of stopped vehicles" refers to the side opposite the vehicle as viewed from the row of stopped vehicles. Road surface information indicates the location of lane markings or road edges.

[0237] Additionally, the curbside parking presence determination unit F51 may be configured to use the intersection's location information to determine whether a stopped vehicle detected by the surrounding monitoring sensors corresponds to a curbside vehicle. Specifically, a vehicle parked on the roadside at the intersection's entrance is more likely to be parked while waiting for a traffic signal, while a vehicle parked on the intersection's exit (i.e., the exit) is less likely to be parked while waiting for a traffic signal. Therefore, whether a parked vehicle is a curbside vehicle may be determined by determining whether the vehicle is parked on the intersection's exit or entrance. For example, a vehicle parked on the curbside at the intersection's entrance may not be considered a curbside vehicle, while a vehicle parked on the curbside at the intersection's exit may be considered a curbside vehicle. Furthermore, the intersection's entrance corresponds to the near side of the intersection.

[0238] As mentioned above, parked vehicles, which are obstacles, are more likely to appear at certain times and locations. Based on past detection history, the map server 2 can also identify / learn the time periods and locations where parked vehicles are more likely to appear, and adjust the parked vehicle detection conditions based on this identification. For example, the conditions for determining the presence of parked vehicles can be relaxed at locations and time periods where these conditions are met. For example, the threshold for the number of vehicles that have taken evasive action can be reduced. Methods for reducing the threshold include setting the threshold to 0. This method allows the presence of parked vehicles to be quickly registered in the roadside parking location DB 251.

[0239] Furthermore, even if a parked vehicle disappears from a location prone to roadside parking, there's a high probability that another vehicle will soon park. Even if a vehicle is replaced, the presence of parked vehicles at that location remains unchanged. Therefore, the conditions for determining the absence of parked vehicles can be set more stringently at locations prone to roadside parking than at other locations. For example, a higher threshold can be set for the number of vehicles that fail to take evasive action. This configuration reduces the risk that the map server 2 will frequently rewrite the registration details of the roadside parking location DB 251 due to changes in parked vehicles.

[0240] In addition, it is difficult for the map server 2 to judge the roadside parking conditions of all general roads at the same time. It is necessary to add priority to the locations that meet the update conditions for processing. For example, parking on the road on the exit side of the intersection has a greater impact on vehicle control. Therefore, it can also be configured to give priority to performing appearance determination processing or disappearance determination processing on locations within a specified distance (for example, 50m) from the intersection exit. In addition, it can also be configured to give priority to performing appearance / disappearance determination from locations on roads with heavy traffic, locations close to stations or large commercial facilities. The priority of updating roadside parking location information on rural areas and roads with average traffic volume below a specified threshold can also be set relatively low. According to such a configuration, the registration content is updated preferentially starting from areas where the usefulness of roadside parking location information is higher, making it easy to ensure real-time performance. As a result, the usefulness of the roadside parking information distribution system 100 can be improved.

[0241] Additionally, the disappearance determination unit G32 can also determine, based on the vehicle status report, whether a vehicle has passed through a roadside parking location without changing its driving position, and determine that a roadside parking vehicle has disappeared based on the presence of a vehicle that has proceeded straight through the roadside parking location. This configuration eliminates the need to transmit a roadside parking location report separately from the vehicle status report. This simplifies vehicle-side processing. In other words, in a configuration that causes each vehicle to transmit a vehicle status report, the contents of the vehicle status report can be referenced as vehicle behavior data, making the roadside parking location report an optional element.

[0242] <Supplementary information on detection and disappearance of parked vehicles on the roadside>

[0243] While the above configuration discloses a configuration in which the vehicle-mounted system 1 uses the front camera 11 to detect parked vehicles, the system is not limited thereto. Alternatively, the system may use side cameras that capture images of the sides of the vehicle or rear cameras that capture images of the rear. Similarly, the system may use side millimeter-wave radars that transmit detection waves to the sides of the vehicle or rear millimeter-wave radars that detect vehicles diagonally behind the vehicle.

[0244] For example, the in-vehicle system 1 or the map server 2 can also use images from side cameras to determine the presence of parked vehicles. If a parked vehicle blocks the first lane, a lane change to the second lane is expected. However, after the lane change, the vehicle is traveling in the second lane, making it less likely that the parked vehicle will be reflected by the front camera 11. As a result, there is a risk that the vehicle will be determined to be free of parked vehicles after the lane change. Using image data from the side camera, where a parked vehicle is located, to determine the presence of parked vehicles can reduce the risk of missing the parked vehicle when passing to the side of the vehicle. Alternatively, the side camera can be a camera installed in a rearview mirror to observe the area to the side. Furthermore, the side cameras and the front camera 11 can be used in a complementary manner. For example, the report data generator F5 can be configured to upload a parked location report containing images captured by the front camera 11 during approach to the parked location and images captured by the side camera after the vehicle changes its driving position.

[0245] In addition, when a vehicle is equipped with multiple cameras, the camera used for identifying parked vehicles on the roadside and the uploaded camera image can be switched according to the surrounding environment of the vehicle. For example, when the distance between the front vehicle and the vehicle is less than a specified threshold value, and the distance between the rear vehicle and the vehicle is above a specified threshold value, the image of the rear camera 11 can be replaced with the image of the rear camera or the side camera as the judgment material for the vehicle-mounted system 1 or the map server 2 on the presence or absence of parked vehicles on the roadside. In addition, when the leading vehicle is a large vehicle such as a truck or a fire truck, and the following vehicle is a small vehicle such as a light car, the rear camera or the side camera can also be used as the camera used to determine the presence or absence of parked vehicles on the roadside. In other words, the cameras used to determine the presence or absence of parked vehicles on the roadside can also be used separately according to whether the field of view in front is wide. The same is true for millimeter-wave radars. When multiple millimeter-wave radars are provided, these multiple millimeter-wave radars can also be used separately according to the surrounding environment.

[0246] Alternatively, devices such as LiDAR or sonar can be used to detect parked vehicles. These devices are also included in perimeter monitoring sensors. Millimeter-wave radar, LiDAR, sonar, etc. can also be referred to as ranging sensors. The map collaboration device 50 can also be configured to use multiple types of perimeter monitoring sensors to detect parked vehicles, etc. For example, the map collaboration device 50 can also detect parked vehicles through sensor fusion.

[0247] The roadside parking presence determination unit F51 or the roadside parking information management unit G3 may also be as follows: Figure 20 The DSM (Driver Status Monitor) 17 detects the driver's seat passenger's intended movements to determine the presence of parked vehicles. The DSM 17 uses a near-infrared camera to capture the driver's seat passenger's face and performs image recognition processing on the captured image, sequentially detecting the driver's seat passenger's facial orientation, gaze direction, eyelid opening, and other factors. For example, the DSM 17 is positioned on the top surface of the steering column cover, the top surface of the instrument panel, or a rearview mirror, with the near-infrared camera facing the driver's seat headrest, enabling it to capture the driver's seat passenger's face.

[0248] For example, the curbside parking presence determination unit F51 or the curbside parking information management unit G3 may determine the presence of a curbside vehicle based on the driver's seat passenger's gaze in the direction of a parked vehicle when passing alongside the curbside vehicle. This is because, if a parked vehicle is present, the driver's seat passenger can be expected to direct their gaze toward the curbside vehicle to confirm that no one has jumped out from behind the parked vehicle. Alternatively, the curbside vehicle may be determined to be absent based on the fact that passengers in vehicles traveling in the second lane are not looking in the direction of the parked vehicle. In other words, the driver's seat passenger's eye movements when passing alongside the curbside vehicle can also serve as information for determining the presence of a curbside vehicle. The in-vehicle system 1 may also upload time-series data on the driver's seat passenger's gaze direction when passing alongside the parked vehicle as a curbside location report. In addition, the in-vehicle system 1 can also upload the result of determining whether the driver's seat passenger's line of sight is directed toward the roadside parking location on the map when passing next to a roadside parked vehicle. The roadside parking information management unit G3 can also determine whether there is a roadside parked vehicle based on the passenger's line of sight information.

[0249] Furthermore, the map collaboration device 50 can upload the behavior of surrounding vehicles to the map server 2 as information for determining the presence of parked vehicles. For example, if a preceding vehicle is detected changing its position to the second lane while traveling in the first lane, the preceding vehicle's change of position to the second lane can be uploaded. Specifically, the front camera 11 can be used to digitize the lane center offset of the preceding vehicle, determine whether a lane change has occurred in front of a parked vehicle on the map, and transmit a parked vehicle location report containing the determination result. Technologies such as SLAM (Simultaneous Localization and Mapping) can be used to determine the behavior of vehicles traveling ahead of the host vehicle. Furthermore, the preceding vehicle is not limited to the preceding vehicle mentioned above; lane changes by subsequent vehicles can also be uploaded. The uploaded behavior of surrounding vehicles is not limited to lane changes; changes in driving position within the first lane can also be uploaded. Furthermore, if the host vehicle is traveling in the second lane, the intrusion from the first lane can be uploaded to the map server 2 as an indicator of the presence of a parked vehicle. Data indicating the behavior of surrounding vehicles is equivalent to other vehicle behavior data. In order to distinguish it from other vehicle behavior data, the vehicle behavior data of the own vehicle will also be referred to as own vehicle behavior data.

[0250] However, suppose a vehicle equipped with the map cooperation device 50, or a mounted vehicle, preemptively changes lanes to the second lane, etc., after receiving roadside parking location information from the map server 2, and then passes alongside the parked vehicle. Therefore, if the map server 2 identifies a roadside parking location, it is unlikely that the mounted vehicle will perform evasive maneuvers near the roadside parking location. If the map server 2 identifies a roadside parking location and begins distributing information about the roadside parking location to each mounted vehicle, the vehicle designated as the target for notification, which performs evasive maneuvers shortly before reaching the roadside parking location, may at best be a vehicle not equipped with the map cooperation device 50, or a non-mounted vehicle. Of course, after the distribution of roadside parking location information begins, the mounted vehicle may perform actions to indicate the presence of the roadside parking location, such as slowing down due to the intrusion of a non-mounted vehicle. However, this deceleration is not always performed to accommodate the intruding vehicle. After the distribution of information about the presence of a roadside parking vehicle begins, the usefulness of the vehicle's behavior data at that location decreases relative to before the distribution begins.

[0251] Based on such a situation, the map cooperation device 50 may also send the behavior data of surrounding vehicles and the detection results of surrounding monitoring sensors as a roadside parking location report in preference to the behavior data of the vehicle itself when it is near a roadside parking location notified in advance from the map server 2. For example, instead of sending the behavior data of the vehicle itself, at least one of the behavior data of surrounding vehicles and the detection results of surrounding monitoring sensors may be sent. Here, the surrounding vehicles that are preferably the subject of the report are other vehicles traveling on the first lane. This is because the first lane is most susceptible to the influence of roadside parked vehicles and is more useful as an indicator of whether roadside parked vehicles remain. According to the above-mentioned structure, the upload of less useful information regarding the disappearance check of roadside parked vehicles can be suppressed. In addition, information that is more useful in the disappearance check of roadside parked vehicles can be preferentially collected to the map server 2.

[0252] Furthermore, even after obtaining roadside parking location information, the vehicle may still drive in the first lane based on the driver's instructions. The map collaboration device 50 may prioritize uploading vehicle behavior data over other vehicle behavior data when the vehicle's driving lane is in the first lane at a predetermined distance from the roadside parking location notified by the map server 2. The map collaboration device 50 may also transmit a dataset containing the vehicle's behavior data as a roadside parking location report if the vehicle's driving lane at the decision point is in the first lane, while transmitting a dataset without the vehicle's behavior data if the vehicle's driving lane is not in the first lane. Specifically, the map collaboration device 50 may transmit a dataset containing the vehicle's behavior data and images captured by the front camera 11 as a roadside parking location report if the vehicle's driving lane at the decision point is in the first lane. On the other hand, if the vehicle's driving lane is not in the first lane when passing the decision point, the map collaboration device 50 transmits a dataset containing the vehicle's behavior data and images captured by the front camera 11 as a roadside parking location report. On the other hand, if the vehicle's driving lane is not in the first lane when passing the decision point, the map collaboration device 50 transmits a dataset containing the behavior of other vehicles traveling in the first lane and detection results from the surrounding monitoring sensors. The decision point can be set, for example, at a location on the vehicle's side that is within a registered roadside parking location report target distance.

[0253] The map coordination device 50 described above corresponds to a configuration that reduces the amount of vehicle behavior data included in a roadside stop report transmitted when the vehicle passes near a received roadside stop, when the vehicle's travel lane at the decision point is not the first lane, compared to when the vehicle is in the first lane. Reducing the amount of vehicle behavior data can be achieved, for example, by increasing the sampling interval or reducing the number of items transmitted as vehicle behavior data. This method of reducing the amount of vehicle behavior information included in the roadside stop report also includes the case where the roadside stop report does not contain any vehicle behavior data. Furthermore, the configuration described above corresponds to a configuration that changes the content of the roadside stop report transmitted based on whether the vehicle is traveling in the first lane when passing the decision point.

[0254] Furthermore, the map cooperation device 50 may be configured to modify the content of the dataset sent to the map server 2 when a parked vehicle not notified by the map server 2 is discovered or when a roadside parking location notified by the map server 2 is passed. For convenience, the dataset sent as a roadside parking location report when a parked vehicle notified by the map server 2 is discovered is also recorded as an unregistered location report. Furthermore, the dataset sent as a roadside parking location report when a roadside parking location notified by the map server 2 is also recorded as a registered location report. For example, an unregistered location report can be a dataset containing the vehicle's behavior data and input data from surrounding monitoring sensors, while a registered location report can be a dataset containing other vehicles' behavior data and input data from surrounding monitoring sensors. A registered location report can be a dataset with the vehicle's behavior data reduced to less than half the size of an unregistered location report. This configuration allows the map server 2 to efficiently aggregate information related to the characteristics of determining the presence and disappearance of parked vehicles.

[0255] Furthermore, when uploading the behavior of surrounding vehicles, there is a possibility that the map server 2 may report the behavior of the same vehicle in an overlapping manner. To prevent the map server 2 from counting the behavior of the same vehicle in an overlapping manner, it is preferable to associate the behavior of the vehicle and surrounding vehicles with their respective vehicle IDs before uploading them. The vehicle IDs of surrounding vehicles can be obtained through inter-vehicle communication or through image recognition of license plates.

[0256] The map collaboration device 50 can also be configured to send information about the vehicle's parking and starting to the map server 2. For convenience, the data set indicating the vehicle's parking is referred to as the vehicle's parking report, and the data set indicating the vehicle's starting is referred to as the vehicle's starting report. The vehicle's parking report is a data set that contains at least information about the location where the vehicle was parked. The vehicle's starting report is a data set that contains at least information about the location where the vehicle was parked, and is equivalent to a data set indicating that the parked vehicle has disappeared from that location. The vehicle's parking report and vehicle's starting report can be understood as a type of roadside parking location report or vehicle status report.

[0257] In a configuration where the map collaboration device 50 does not transmit a parking report, even if the vehicle is parked on the road, the map server 2 cannot recognize this until another vehicle detects this and transmits a roadside parking location report related to the vehicle to the map server 2. In contrast, in a configuration where the map collaboration device 50 spontaneously transmits a parking report, the map server 2 can identify the roadside parking location resulting from the vehicle's movement in real time. Furthermore, in a configuration where the map collaboration device 50 transmits a vehicle start report, the map server 2 can identify the location where the vehicle, which was parked on the roadside, has disappeared in real time, without waiting for reports from other vehicles.

[0258] Furthermore, for example, it is possible to determine that the vehicle is parked based on the vehicle speed reaching zero. Alternatively, it is possible to determine that the vehicle is parked based on the fact that the driving power source is turned off. It is possible to detect that the vehicle has started based on the fact that the vehicle speed reaches or exceeds a predetermined threshold (e.g., 10 km / h) after the driving power source is turned on. Various methods can be used to determine whether the vehicle is parked or started.

[0259] The roadside parking presence / absence determination unit F51 may be configured to determine whether Figure 21 The sequence shown reports to the map server 2 that the vehicle has parked in a gap in the existing parking row. Figure 21 The flowchart shown includes steps S801 to S805 as an example. As mentioned above, the number of steps and the processing order of the flowchart can be changed as appropriate. The above-mentioned various processing can be carried out in parallel, in combination, or in substitution. Figure 21 The processing flow shown in FIG. can be performed periodically while the vehicle is powered on. Figure 21 The processing shown.

[0260] First, in step S801, the detected object information acquisition unit F4 acquires sensing information from the surrounding monitoring sensors and moves to step S802. In step S802, based on the sensing information acquired in step S801, it is determined whether a plurality of parked vehicles arranged along the roadside, i.e., a row of parked vehicles, is detected. If a row of parked vehicles is detected, step S802 is affirmatively determined and the process proceeds to step S803. On the other hand, if a row of parked vehicles is not detected, step S802 is negatively determined and the process proceeds to step S803. In addition, the roadside parking information acquisition unit F21 can also use the roadside parking location information acquired from the map server 2 to determine whether the parked vehicle corresponds to a roadside parked vehicle.

[0261] In step S803, the report data generator F5 detects gaps, or gaps, between parked vehicles on the roadside based on the sensing information acquired in step S801. Alternatively, the report data generator F5 may detect gaps based on the roadside parking location information acquired by the roadside parking information acquisition unit F21, in lieu of sensing information or in a complementary manner. Detecting gaps is equivalent to identifying areas where gaps exist. A gap is defined as a vacant space at least long enough for the vehicle to perform parallel parking. For example, a vacant space of at least 2 meters along a roadside edge is considered a gap. If no gaps exist in step S803, the process can be terminated.

[0262] In step S804, the report data generator F5 determines whether the vehicle is parked in the gap detected in step S803. For example, it can be determined that the vehicle is parked in a gap based on the vehicle speed being zero within the gap. If the vehicle is parked in the gap, the report data generator F5 generates a parking report and transmits it to the map server 2 (step S805). If the vehicle is not parked in the gap detected in step S803, for example, if the vehicle passes to the side of the gap, the process ends. Furthermore, the above process can be performed as the vehicle moves or at a constant time.

[0263] As described above, by configuring the map cooperation device 50 to transmit a parking report, the map server 2 can identify the location of a vehicle parked on the road in real time. In particular, by configuring the parking report to be transmitted based on the vehicle parking in a gap in a row of roadside parking areas, the map server 2 can process the extension / joining and disconnection of roadside parking areas, which will be described later, in real time.

[0264] In addition, the above example illustrates the situation where the vehicle is parked in a gap in an existing parking row, but the scenario for generating and sending the vehicle parking report is not limited to this. It is not limited to the gap in a roadside parking row, and the vehicle parking report can also be sent based on the vehicle being parked on the road. The map cooperation device 50 can also generate and send the vehicle parking report based on, for example, the vehicle being parked along the road edge. Parking along the road edge refers to a parking method in which the distance from the road edge to the side of the vehicle is within the parking determination distance, and the angle between the front and rear directions of the vehicle and the road edge is less than a specified value (for example, 30 degrees). The distance from the road edge to the side of the vehicle and the orientation of the vehicle relative to the road edge can be determined based on the inspection results of the surrounding monitoring sensors. The vehicle start report can also be sent not only when exiting the parking row, but also at all times when starting. In addition, it can be configured so that the vehicle parking report and vehicle start report are sent based on the fact that the vehicle is parked on the road.

[0265] <Calculation of Detection Reliability by Map Cooperation Device 50>

[0266] The roadside parking presence determination unit F51 may also calculate the possibility of the actual presence of a roadside parked vehicle as the detection reliability based on a combination of whether the vehicle is detected by the front camera 11, whether the vehicle is detected by the millimeter wave radar 12, and whether an evasive action is taken. Figure 22 The figure may also be constructed so that the more points (sensors, actions, etc.) that indicate the presence of parked vehicles on the roadside, the higher the detection reliability is calculated. Figure 22 The method of determining the detection reliability shown is an example and can be modified as appropriate.

[0267] also, Figure 22 The vehicle behavior when the vehicle is traveling in the first lane refers to the avoidance action of the vehicle. When the vehicle is traveling in the second lane, the behavior of the surrounding vehicles traveling in the first lane can be used instead for the calculation of the detection reliability. For example, the presence or absence of a lane change (that is, insertion) from the first lane to the second lane can be used as a perspective for calculating the detection reliability. In addition, when there is an insertion from the first lane to the second lane, it can be expected that the traffic flow in the second lane will also slow down. Therefore, when the vehicle is traveling in the second lane, it can also be determined that the surrounding vehicles are taking avoidance actions when a reduction in driving speed is observed near the roadside stop on the map.

[0268] The detection reliability calculated by the roadside parking presence determination unit F51 may also be included in the roadside parking location report. The map server 2 may also statistically process the detection reliability included in the reports from multiple vehicles to determine whether a roadside vehicle is present. Detection reliability may also be evaluated using passenger line-of-sight information detected by, for example, a DSM. For example, if the driver's seat passenger's line of sight is directed in the direction of a roadside parked vehicle while passing alongside the vehicle, the detection reliability may be set higher.

[0269] The above-mentioned detection reliability indicates the reliability of reports that there are parked vehicles on the roadside. Therefore, the above-mentioned detection reliability can also be called existence report reliability. In addition, the roadside parked vehicle presence determination unit F51 can also calculate the possibility that there are no parked vehicles on the roadside as non-detection reliability based on a combination of whether it is detected by the front camera 11, whether it is detected by the millimeter-wave radar 12, and whether there is avoidance action. The non-detection reliability is equivalent to the inverse of the above-mentioned detection reliability. The higher the detection reliability, the lower the non-detection reliability can be set. The non-detection reliability indicates the reliability of reports that there are no parked vehicles on the roadside. Therefore, the above-mentioned non-detection reliability can also be called non-existence report reliability.

[0270] Accuracy of detecting parked vehicles on the roadside

[0271] The map server 2 may also be configured to calculate the probability of the presence of a parked vehicle on the roadside as the actual presence accuracy and distribute it. The actual presence accuracy is equivalent to the reliability of the determination result and notification of the presence of a parked vehicle on the roadside. For example, the roadside parking information management unit G3 may also be configured as follows: Figure 23 The vehicle shown includes an accuracy calculation unit G33 for calculating the reliability of the determination result of the presence of a roadside parked vehicle as actual presence accuracy.

[0272] The accuracy calculation unit G33 calculates the actual presence accuracy based on the behavior data of multiple vehicles and the ratio of vehicles that have taken avoidance actions. Figure 24 As shown, the greater the number of vehicles that report the presence of parked vehicles, the higher the actual presence accuracy is set. Vehicles that report the presence of parked vehicles include not only vehicles that have changed their driving position from the first lane to the second lane, but also vehicles that are traveling in the second lane and have uploaded detection results of parked vehicles. In addition, the accuracy calculation unit G33 may set the case where the presence of parked vehicles has been confirmed through image analysis based on the server processor 21 or visual observation by the operator as 100, and calculate the actual presence accuracy based on the number and type of reports indicating the presence of parked vehicles. For example, the greater the number of vehicles that have implemented avoidance actions or the number of vehicles whose surrounding monitoring sensors have detected parked vehicles, the higher the actual presence accuracy may be set.

[0273] The accuracy calculation unit G33 may also perform calculations based on the difference between the number of reports of parked vehicles and the number of reports of no parked vehicles. For example, if the number of reports of parked vehicles and no parked vehicles is equal, the actual presence accuracy may be 50%. The accuracy calculation unit G33 may also perform statistical processing on the detection reliability contained in reports from multiple vehicles to calculate the actual presence accuracy. The accuracy calculation unit G33 may also periodically calculate the actual presence accuracy.

[0274] The distribution processing unit G4 may also distribute a roadside parking location notification packet containing the aforementioned actual presence accuracy. If the actual presence accuracy of a roadside parking vehicle at a particular location changes, the distribution processing unit G4 may also distribute a roadside parking location notification packet containing the updated actual presence accuracy to vehicles that have already been distributed a roadside parking location notification packet for that location. For example, the distribution processing unit G4 may also periodically distribute roadside parking location notification packets along with information containing the probability of the roadside parking vehicle. The distribution processing unit G4 may also distribute roadside parking location notification packets at regular intervals, indicating the actual presence accuracy in three levels: "still present," "high probability still present," and "high probability no longer present."

[0275] The value obtained by subtracting the actual presence accuracy from 100% corresponds to the disappearance accuracy indicating the probability of the parked vehicle disappearing. The distribution processing unit G4 may also transmit a disappearance notification packet including the disappearance accuracy of the parked vehicle.

[0276] <Utilization of Commercial Vehicle Operation Information>

[0277] Commercial vehicles such as buses, taxis, trucks, and moving vehicles can also be configured to report parking and movement to the map server 2 in real time. Commercial vehicles can include construction vehicles. For example, the above-mentioned commercial vehicles can also be configured to be equipped with a vehicle-mounted device for commercial vehicles that is a map cooperation device 50 configured to report parking positions and movement status, so that the map server 2 can grasp its parking position, etc. In such a system configuration, when the parking position obtained from the above-mentioned vehicle-mounted device for commercial vehicles is on the road, the map server 2 registers the location as a roadside parking location. In addition, when a report on the purpose of resuming movement is received from the vehicle-mounted device for commercial vehicles, the setting of the roadside parking location corresponding to the vehicle is canceled.

[0278] The distribution processing unit G4 of the map server 2 can also set the actual existence accuracy of roadside parking location information based on parking reports from onboard devices for commercial vehicles to a high level and distribute it immediately. Furthermore, the distribution processing unit G4 can also set the disappearance accuracy of disappearance notification packets based on start reports from onboard devices for commercial vehicles to a high level and distribute them immediately. This is because all information is based on reports from the parties involved and has a higher reliability. Furthermore, the map server 2 can also indirectly obtain the parking location and movement resumption information of the commercial vehicles mentioned above from the operating company / operation management server that manages the operation of the vehicles. Furthermore, the configuration for transmitting parking and start information to the map server 2 can also be applied to map cooperation devices 50 for vehicles other than commercial vehicles, such as private cars.

[0279] <Supplementary information on the distribution of roadside parking information>

[0280] The roadside parking location notification packet preferably includes the location, type, and size of the roadside parked vehicle. Alternatively, the roadside parked vehicle location information may include not only the location coordinates but also the amount of protrusion into the first lane as detailed location information. Furthermore, the roadside parking location notification packet may include information on the width of the area in the first lane where the vehicle can travel, excluding the area blocked by the roadside parked vehicle.

[0281] Because the roadside parking location notification packet includes the vehicle's protrusion amount and the permitted travel width, vehicles receiving the roadside parking location notification packet can determine whether a lane change is necessary or whether lateral position adjustment is feasible. Furthermore, when traveling across a lane dividing line, the vehicle's protrusion amount into the second lane can be calculated. By calculating the protrusion amount into the second lane, inter-vehicle communication can be used to notify vehicles traveling in the second lane of the vehicle's protrusion amount, enabling coordinated driving positions with surrounding vehicles.

[0282] Furthermore, the roadside parking location notification data packet may include information on the time when a roadside parking vehicle was determined to have occurred, as well as the latest (i.e., final) time when the roadside parking vehicle was determined to still exist. By including these determination times, the vehicle receiving the information can estimate the reliability of the received information. For example, the shorter the elapsed time from the final determination time, the higher the reliability. The roadside parking location notification data packet may also include information such as the number of vehicles that have confirmed the presence of the roadside parking vehicle. The greater the number of vehicles that have confirmed the presence of the roadside parking vehicle, the higher the reliability of the roadside parking information can be estimated. Furthermore, the reliability of the roadside parking information may be adjusted to determine whether the information is used for vehicle control or limited to passenger notification.

[0283] The roadside parking location notification data packet may also include characteristic information such as the color tone of parked vehicles. Alternatively, it may include an image of the parked vehicle captured by a vehicle. This configuration makes it easier for the in-vehicle system 1 or a passenger planning to pass through the roadside parking location to associate parked vehicles notified by the map server 2 with real-world parked vehicles. As a result, the accuracy of determining whether a parked vehicle notified by the map server 2 is still present or gone is improved.

[0284] The distribution processing unit G4 may also set and distribute a lane change recommendation POI (Point of Interest) at a location within the lane of a parked vehicle, a predetermined distance before the roadside parking location. A lane change recommendation POI is a location where a lane change is recommended. By setting and distributing the lane change recommendation POI in the map server 2 in this manner, the vehicle can omit the process of calculating the lane change point, reducing the processing load on the processing unit 51 and the driving assistance ECU 60. In a configuration that prompts the user to change lanes, the lane change recommendation POI can also be used to determine the timing for displaying the parked vehicle notification image.

[0285] The roadside parking location notification packet may also include information indicating whether a risk remains at the location, such as whether the parked vehicle has disappeared. The presence accuracy described above may also be used to indicate whether a risk remains at the location. Similar to the roadside parking location notification packet, the roadside vehicle disappearance packet preferably includes information such as characteristics of the roadside vehicle and the time at which disappearance was determined.

[0286] Alternatively, the distribution processing unit G4 may be configured to distribute roadside stop location notification packets only to vehicles executing a specified application, such as an autonomous driving application. Specified applications may include, in addition to autonomous driving applications, ACC (Adaptive Cruise Control), LTC (Lane Trace Control), and navigation applications. Furthermore, in a configuration that pulls roadside stop location information, the map cooperation device 50 may be configured to conditionally request roadside stop location information from the map server 2 based on the execution of a specific application. This configuration can suppress excessive information distribution and improve the control stability of the driving assistance ECU 60. Furthermore, the distribution processing unit G4 may be configured to push-distribute roadside stop location notification packets only to vehicles that have been set to automatically receive roadside stop location information based on user settings. This configuration reduces the risk of wireless communication between the map server 2 and the map cooperation device 50 without the user's intent.

[0287] Furthermore, the distribution processing unit G4 can also distribute roadside parking location information on a grid / map tile basis. For example, the roadside parking location information on a map tile can be distributed to vehicles located in that map tile or to vehicles requesting a map for that map tile. This configuration simplifies the selection of distribution targets and allows for the simultaneous distribution of information on multiple roadside parking locations. Consequently, the processing load on the map server 2 can be reduced. Furthermore, how the received roadside parking location information is used depends on which application is active in the in-vehicle system 1. This configuration increases the diversity and flexibility of the uses of roadside parking location information in the in-vehicle system 1.

[0288] <Upload processing of the in-vehicle system 1>

[0289] The map collaboration device 50 can also be configured to send a roadside parking location report only when the content registered on the map as roadside parking information differs from the content observed by the vehicle. In other words, it can also be configured not to send a roadside parking location report when the map content matches the actual situation. For example, a roadside parking location report can be sent when a roadside parking vehicle is observed at a location where the presence of a roadside parking vehicle is not registered on the map, or when a roadside parking location report is not observed at a location where a roadside parking vehicle is registered on the map. The above-mentioned configuration can suppress communication volume. In addition, the server processor 21 can not perform determination processing related to the presence of roadside parking vehicles on the portion of the real world that matches the content registered on the map. In other words, the processing load of the server processor 21 can also be reduced.

[0290] Furthermore, while the configuration disclosed herein involves the map collaboration device 50 spontaneously uploading vehicle behavior data to the map server 2 when a vehicle passes near a parked vehicle, the configuration of the map collaboration device 50 is not limited to this. Alternatively, the map collaboration device 50 can upload vehicle behavior data to the map server 2 only when a vehicle performs a specified action, such as a lane change or sudden deceleration. In a configuration where vehicle behavior data is uploaded only when a vehicle performs a specific action, there is a concern that it may be difficult for the map server 2 to collect information needed to determine whether a parked vehicle has disappeared. This is because if a parked vehicle disappears, the vehicle does not perform any specific action.

[0291] To address this concern, the server processor 21 can also transmit a control signal, or upload instruction signal, to vehicles currently passing or scheduled to pass a roadside stop on the map, instructing them to upload a roadside stop report. In other words, the map collaboration device 50 can also be configured to determine whether to upload a roadside stop report based on instructions from the map server 2. This configuration allows the map server 2 to determine the status of each vehicle's roadside stop report upload, suppressing unnecessary communication. For example, if sufficient information regarding the presence and absence of parked vehicles can be collected, suppressing vehicle uploads is also possible.

[0292] Additionally, the server processor 21 can also set a location where vehicle activity indicating the presence of a parked vehicle is observed based on a vehicle status report as a verification location and transmit an upload instruction signal to vehicles scheduled to pass through the verification location. For example, a location where vehicle activity indicating the presence of a parked vehicle is observed could be a location where two or three vehicles have successively changed lanes. This configuration allows for centralized and rapid collection of information on locations suspected of having parked vehicles, enabling real-time detection of the presence of parked vehicles.

[0293] Furthermore, the vehicle can be configured to configure whether to upload roadside parking location reports. For example, the user can configure whether to upload roadside parking location reports via an input device. Furthermore, the user can also configure and change the information items uploaded as roadside parking location reports. This configuration can reduce the concern that users may inadvertently upload vehicle behavior data to the map server 2, thereby increasing communication traffic. Furthermore, from the perspective of privacy protection, the transmission source information can be rewritten using a specified encryption code to a number different from the actual vehicle ID and uploaded to the map server 2.

[0294] Furthermore, the roadside parking information distribution system 100 can be configured to reward users who actively upload information related to parked vehicles. By providing incentives for sending roadside parking location reports, it is easier to collect information related to parked vehicles, thereby improving the effectiveness of the roadside parking information distribution system 100. Incentives can include reduced vehicle taxes, reduced map service usage fees, or points that can be used to purchase items or services. Points that can be used to purchase specified items or services also include the concept of electronic currency.

[0295] <Grouping of roadside stops>

[0296] As mentioned above, there are locations and time periods where parked vehicles are easily parked. Therefore, even if a location initially has only one parked vehicle, new parked vehicles may appear before and after it over time, eventually filling up the remaining area. Based on this scenario, the map server 2 can also be configured to group locations where multiple parked vehicles are concentrated into roadside parking areas and distribute them accordingly. For example, the map server 2 can provide point-based notifications when there is only one parked vehicle, but if other vehicles are detected parked before and after it, define these areas as roadside parking areas and distribute them accordingly. Figure 25 This is a diagram showing the configuration of the map server 2 corresponding to this technical concept, and includes a grouping unit G34.

[0297] The grouping unit G34 is a structure for grouping a plurality of roadside parking locations according to a prescribed rule. The grouped roadside parking locations are equivalent to the roadside parking areas described above. The grouping unit G34, for example, groups roadside parking locations whose intervals in the direction of road extension are less than a prescribed threshold value (hereinafter referred to as the connection distance) into one group. The connection distance can be, for example, 25m, 50m, 100m, 200m or 250m. If the connection distance is short, the roadside parking area is refined, for example, the autonomous driving vehicle is frequently urged to change lanes. Therefore, it is desirable that the connection distance used as the threshold for grouping has a certain length. In addition, the connection distance can also be changed according to the type of road (national highway, prefectural road, etc.) and the speed limit. The connection distance can also be set longer as the road scale is larger and the speed limit is higher.

[0298] In addition, roadside parking locations that are separated from the roadside parked vehicles in front and behind by more than the connection distance are treated as other groups or points. Figure 26 As shown, if the connection distance is set to 100 meters, for example, parked vehicles Pv1 through Pv3, which are spaced less than 100 meters apart, are considered to be part of the same group Gr1. Meanwhile, parked vehicle Pv4, which is spaced more than 100 meters apart from parked vehicle Pv1, is treated as a separate group or point. A point in this context refers to an individual parked vehicle that is not included in any group (i.e., a roadside parking area).

[0299] In addition, the grouping unit G34 can also be used even if the distance between the grouping unit G34 and the roadside parked vehicles in front and behind is within the connection distance. Figure 27As shown, when there is a connection point with another road (hereinafter referred to as branch road R1) in the middle, the connection point is also used as a node and divided into different groups Gr2 and Gr3. By dividing the roadside parking area with the connection point with the branch road R1 as a node in this way, it is possible to prevent vehicles from entering the road due to roadside parking area information. In addition, the map server 2 can also be configured to ignore the branch road R1 and group multiple roadside parked vehicles when it is configured to obtain path planning from each vehicle-mounted system 1 and when there is no vehicle planning to enter the path of the branch road R1. For example, groups Gr2 and Gr3 can also be merged into one group. In other words, multiple roadside parking locations that exist via the branch road R1 can also be defined as one roadside parking area.

[0300] The grouping unit G34 updates the size of the roadside parking area (and thus updates the group structure) as roadside parking spots appear and disappear. Figure 28 As shown, if a roadside parking area (group Gr4) is empty due to the disappearance of parked vehicles Pv5 and Pv6, resulting in a gap greater than the connecting distance within the roadside parking area, the area is divided into two roadside parking areas (groups Gr5 and Gr6). This configuration reduces the risk of being unable to park near the roadside shoulder based on outdated roadside parking area information. Furthermore, in a configuration where each map coordination device 50 transmits a vehicle parking report and a vehicle start report, the grouping unit G34 can update the roadside parking areas in real time based on these reports.

[0301] In the configuration including the grouping unit G34 as described above, the distribution processing unit G4 can also collectively distribute information on multiple roadside parking locations belonging to the same group as roadside parking area information. Roadside parking area information includes the coordinates of the starting location of the area and its distance. Roadside parking area information can also include information such as the number of parked vehicles forming the roadside parking area, the location information of the end of the area, and the direction in which the roadside parking area extends. Furthermore, the roadside parking area information can also include characteristics (color, type) of the vehicles at the beginning and end of the area. If the roadside parking area information includes characteristic information of the first and last vehicles, the map coordination device 50 can easily verify the current status of the location indicated by the roadside parking area information. For example, it can easily verify whether the roadside parking area is expanding or contracting. Furthermore, by collectively distributing information on multiple roadside parking locations as roadside parking area information, overall communication traffic can be reduced. The map coordination device 50 no longer needs to process multiple roadside parking locations independently, thereby reducing the processing load on the map coordination device 50.

[0302] Furthermore, the map collaboration device 50 may limit the roadside parking location reports uploaded when traveling in locations defined as roadside parking areas to data near the starting and ending points of the parking areas. This is because, in the case of long parking areas, continuously uploading image data while traveling within that area may increase communication costs and traffic. However, there is also the possibility that vacant space exceeding the connecting distance may occur within the area designated as a roadside parking area due to the movement of parked vehicles. Alternatively, the system may upload image data whenever traveling through a roadside parking area, for example, each time the vehicle travels the connecting distance or half the connecting distance.

[0303] The distribution processing unit G4 notifies roadside parking locations that do not belong to any roadside parking area or group as points. Furthermore, if the number of vehicles constituting a roadside parking area decreases and point processing is adopted, in other words, if the grouping is terminated, the distribution of roadside parking area information can be switched to point notification. Roadside parking area information, which aggregates information on multiple roadside parking locations, is also included in the concept of roadside parking location information.

[0304] <Example of citing information on a group of vehicles that are parked>

[0305] The above discloses a method for detecting the position of a vehicle parked at the end of a general road in real time based on the behavior data of vehicles passing around the vehicle and the detection results of the surrounding monitoring sensors. However, the above-mentioned method for detecting the position of a parked vehicle can also refer to a technology for detecting the position information of the end of a traffic jam in real time. Figure 29 as well as Figure 30 An example of a traffic congestion information distribution system that determines and distributes traffic congestion sections based on reports from vehicles will be described.

[0306] The map cooperation device 50 constituting the traffic congestion information distribution system of the present disclosure is as follows: Figure 29 As shown, in addition to the vehicle position acquisition unit F1, the map server 2 also includes a traffic congestion information acquisition unit F22. Figure 30 As shown, in addition to the report data acquisition unit G1 and the distribution processing unit G4, the traffic congestion information management unit G5 and the traffic congestion point DB 27 are provided. The traffic congestion information management unit G5 includes a generation determination unit G51 and an end determination unit G52.

[0307] The traffic congestion information acquisition unit F22 is configured to acquire the location of a traffic congestion zone or its end from the map server 2. Based on the information acquired by the vehicle behavior acquisition unit F3 and the detected object information acquisition unit F4, the report data generation unit F5 uploads data indicating the current status of the traffic congestion zone, or traffic congestion notification zone, notified from the map server 2 as a traffic congestion zone report to the map server 2. The content uploaded as the traffic congestion zone report may include vehicle behavior data while traveling through the traffic congestion notification zone or image data captured within the traffic congestion notification zone. Alternatively, it may include positional information of parked vehicles detected by surrounding monitoring sensors. Furthermore, the report may include multiple types of information as described above.

[0308] The traffic congestion information management unit G5 updates the location and length of the end of a traffic congestion based on traffic congestion zone reports received from vehicles. If a vehicle is parked at a location that was previously determined not to be a traffic congestion zone, the occurrence determination unit G51 sets that location as a traffic congestion zone. The method for detecting the occurrence of a traffic congestion location can be the same as the method for determining the presence of roadside parking locations. Furthermore, it is sufficient to determine whether a line of parked vehicles on the road is a line based on traffic congestion or a line based on parked vehicles on the road based on the lateral distance from the road end and the lane ID. For example, a line of vehicles that is at least a specified distance (e.g., 0.5 m) from the road end, or a line of vehicles in the second or third lane, can be determined as a line based on traffic congestion.

[0309] The end determination unit G52 is configured to update the location of the end of a traffic jam based on reports from vehicles. For example, if a vehicle passes through a location previously determined as the end of a traffic jam at a speed exceeding a predetermined threshold, the end determination unit G52 shortens the traffic jam section and recalculates the end of the traffic jam. The change in the end of a traffic jam can employ the same determination logic used to determine the presence or absence of parked vehicles on the roadside. Furthermore, the vehicle behavior used to detect a traffic jam section is not limited to a change in driving position from the first lane to the second lane, as used in detecting parked vehicles on the roadside. Lane changes from any lane to another lane can be used as vehicle behavior data for determining the end of a traffic jam. Furthermore, stopping at a location more than a predetermined distance from a traffic light, the activation of hazard lights, and other vehicle behavior data (i.e., judgment data) for determining the end of a traffic jam can also be used. While traffic congestion conditions, such as the presence or absence of a traffic jam and the location of the end of a traffic jam, are determined on a lane-by-lane basis, alternative methods may also be employed to determine the end of a traffic jam.

[0310] The traffic congestion location database 27 stores information on traffic congestion zones generated by the traffic congestion information management unit G5. The traffic congestion zone data stored in the traffic congestion location database 27 is updated regularly by the traffic congestion information management unit G5. The traffic congestion information management unit G5 updates the traffic congestion zones sequentially, starting with the location where the number of vehicle reports exceeds a predetermined threshold. The distribution processing unit G4 distributes the data stored in the traffic congestion location database 27 to vehicles.

[0311] The above configuration enables lane-by-lane determination of traffic congestion conditions, including the presence or absence of traffic congestion and the location of the end of the congestion. This configuration allows for lane-by-lane traffic congestion determination, distinguishing between traffic congestion involving right and left turns and overall road congestion, and distributing these information accordingly. Vehicles planning to proceed straight through an intersection can take countermeasures, such as changing lanes, by detecting traffic congestion involving right and left turns ahead of their lane.

[0312] The above configuration enables information on traffic congestion sections, particularly real-time position information at the end of a traffic congestion, to be distributed to each vehicle. By acquiring real-time position information at the end of a traffic congestion, each onboard system 1 can execute vehicle control such as lane changes and deceleration with sufficient time and distance.

[0313] Application examples for autonomous driving

[0314] The roadside parking information and traffic congestion information of each lane generated by the map server 2 can also be used, for example, to determine whether autonomous driving can be performed. As a road condition for autonomous driving, there can also be a configuration in which the number of specified lanes is greater than a specified amount n. The specified amount n is an integer greater than "2", for example, "2", "3", "4", etc. In such a configuration, an interval in which the number of valid lanes is less than n due to traffic congestion caused by parked vehicles on the roadside or waiting to turn right or left can become an interval in which autonomous driving is not possible. The number of valid lanes refers to the number of lanes in which the vehicle can actually travel. For example, if one lane of a two-lane road is blocked by a parked vehicle on the roadside, the number of valid lanes of the road is "1".

[0315] It can also be configured to determine whether it corresponds to a non-automatic driving section on the vehicle side (for example, the driving assistance ECU 60, the automatic driving ECU). In addition, the map server 2 can also set a non-automatic driving section based on the roadside parking information and distribute the non-automatic driving section. For example, in the map server 2, the section where the number of effective lanes is insufficient due to the roadside parked vehicle is set as a non-automatic driving section and distributed, and when the disappearance of the roadside parked vehicle is confirmed, the non-automatic driving setting is released and distributed. In addition, the server that distributes the setting of the non-automatic driving section can also be used as the automatic driving management server 7, such as Figure 31 As shown, it is set up independently from the map server 2. The automatic driving management server is equivalent to a server that manages the sections where automatic driving is possible or not. As mentioned above, the roadside parking information can be used to determine whether the operational design area (ODD: Operational Design Domain) set for the vehicle is met. Figure 31 As shown in FIG, a system that distributes information related to whether automatic driving is possible to vehicles based on roadside parking information is called a non-automatic driving section distribution system.

[0316] <Postscript (1)>

[0317] The control unit and method described in the present disclosure can also be implemented by a dedicated computer comprising a processor programmed to execute one or more functions embodied by a computer program. Furthermore, the apparatus and method described in the present disclosure can also be implemented by dedicated hardware logic circuits. Furthermore, the apparatus and method described in the present disclosure can also be implemented by one or more dedicated computers comprising a combination of a processor executing a computer program and one or more hardware logic circuits. Furthermore, a computer program can also be stored as instructions executed by a computer on a non-migratable, tangible recording medium that can be read by a computer. For example, the units and / or functions provided by the map collaboration device 50 and the map server 2 can be provided solely through software, solely through hardware, or through a combination thereof, using software stored in a physical memory device and a computer executing the software. Alternatively, some or all of the functions provided by the map collaboration device 50 and the map server 2 can be implemented as hardware. Implementing a function as hardware includes implementing it using one or more integrated circuits. For example, the server processor 21 can be implemented using an MPU or GPU instead of a CPU. Furthermore, the server processor 21 can be implemented by combining multiple types of computing processing devices, such as CPUs, MPUs, and GPUs. Furthermore, the ECU can also be implemented using an FPGA (field-programmable gate array) or an ASIC (application-specific integrated circuit). The same applies to the processing unit 51. Various programs can be stored in a non-transitory tangible storage medium. As a storage medium for the program, various storage media such as an HDD (Hard-disk Drive), an SSD (Solid State Drive), an EPROM (Erasable Programmable ROM), a flash memory, a USB memory, and an SD (Secure Digital) memory card can be used.

[0318] <Postscript (2)>

[0319] The present disclosure also includes the following components. The map server described below corresponds to the parking location management device, the appearance determination unit corresponds to the parking location detection unit, and the disappearance determination unit corresponds to the presence / absence determination unit. Furthermore, the information presentation control device corresponds to the HMI system / HCU. ​​The driving assistance device corresponds to the driving assistance ECU.

[0320] The conditions for determining the presence of a roadside parked vehicle differ between the appearance determination unit and the disappearance determination unit of the map server.

[0321] The map server is configured to acquire an image obtained by capturing a location where a parked vehicle is located from a vehicle, and the disappearance determination unit determines that a parked vehicle has moved based on the number or ratio of vehicles not performing evasive action at the location exceeding a threshold. The threshold used to determine that a parked vehicle has disappeared is changed based on whether the parked vehicle is reflected in the image.

[0322] The map server is configured to instruct a vehicle scheduled to pass through a roadside parking point to transmit a predetermined type of information including vehicle behavior data for determining the presence status of the roadside parked vehicle.

[0323] The map server is configured such that at least one of the appearance determination unit and the disappearance determination unit uses camera images captured by the vehicles in addition to vehicle behavior data of the plurality of vehicles to determine whether a vehicle is parked on the roadside.

[0324] The map server is configured to change a combination of information types used to determine the presence of a parked vehicle when determining the presence and disappearance of a parked vehicle.

[0325] The map server is configured to use the analysis results of the image captured by the vehicle-mounted camera when determining appearance, but not to use the analysis results of the image captured by the vehicle-mounted camera when determining disappearance.

[0326] The map server is configured to change the weight of each type of information used to determine the presence of a parked vehicle when determining the presence and disappearance of a parked vehicle.

[0327] The map server is configured to use analysis results of images captured by an onboard camera as information for determining the presence of parked vehicles on the roadside, and to weight the image analysis results less heavily when determining disappearance than when determining presence.

[0328] The map server is configured to determine the presence or absence of parked vehicles on the roadside by comparing the traffic volume for each lane.

[0329] The map server is configured to use lane changes performed after deceleration as avoidance actions. This configuration also eliminates lane changes for overtaking.

[0330] The map server is configured not to distribute roadside parking location information to a vehicle that is traveling or scheduled to travel in a lane that is more than one lane away from the first lane.

[0331] Even if a parked vehicle is detected by the camera, the roadside parking presence determination device or the map server does not determine that a parked vehicle exists unless a three-dimensional object is detected by the distance measuring sensor.

[0332] The map cooperation device as a vehicle device is configured to upload a roadside stop report including vehicle behavior to the map server based on an instruction from the map server or voluntarily when the vehicle is traveling within a predetermined range from a roadside stop notified from the map server.

[0333] The map cooperation device includes a report processing unit that, when passing near a roadside stop acquired from a map server, transmits a data set including at least one of vehicle behavior data representing the behavior of the vehicle, other vehicle behavior data representing the behavior of other vehicles, and an image from an onboard camera as a roadside stop report to the server.

[0334] The map cooperation device is configured such that the report processing unit changes the content of the roadside stop report sent to the server based on whether the vehicle's driving lane at a determination point a predetermined distance before the notified roadside stop is a lane adjacent to a road edge.

[0335] The map cooperation device outputs the roadside parking location information obtained from the map server to the navigation device or the autonomous driving device.

[0336] The information presentation control device displays on the display a roadside parked vehicle notification image generated based on roadside parked location information acquired from the map server.

[0337] When the vehicle is traveling / planning to travel in a lane that is more than one lane away from the first lane, the information presentation control device does not notify passengers of information about parked vehicles on the roadside.

[0338] The driving assistance device is configured to switch, based on the accuracy of the actual presence of parked vehicles notified from the map server, whether to execute vehicle control based on the information or to provide only information.

Claims

1. A parking location management device, wherein: have: a determination material acquisition unit that acquires at least one of vehicle behavior data indicating the behavior of the vehicle and sensing information from a periphery monitoring sensor mounted on the vehicle in correspondence with position information of at least one vehicle; a parking location detection unit that detects a location where a vehicle is parked on a general road, i.e., a roadside parking location, based on the information acquired by the determination material acquisition unit; a presence state determination unit for determining whether a vehicle is still present at the roadside parking location detected by the parking location detection unit based on the information acquired by the determination material acquisition unit; as well as a distribution processing unit that distributes information about the roadside parking location detected by the parking location detection unit to the vehicle, The parking location detection unit is configured as follows: Based on the fact that the distance between the vehicle stopped on the road and the road end is within a predetermined distance, the location where the vehicle is located is set as the roadside parking location, and The roadside parking spot on the road on the exit side of the intersection is detected with priority over the roadside parking spot on the road on the entrance side of the intersection.

2. A parking location management device, wherein: have: a determination material acquisition unit that acquires at least one of vehicle behavior data indicating the behavior of the vehicle and sensing information from a periphery monitoring sensor mounted on the vehicle in correspondence with position information of at least one vehicle; a parking location detection unit that detects a location where a vehicle is parked on a general road, i.e., a roadside parking location, based on the information acquired by the determination material acquisition unit; as well as The existence state determination unit determines whether there is still a vehicle at the roadside parking location detected by the parking location detection unit based on the information acquired by the determination material acquisition unit, The parking location detection unit is configured to detect the location where the parked vehicle is located as the roadside parking location based on the fact that there is no intersection or traffic light within a first distance specified on the road travel direction side from the parked vehicle and the distance from the road end to the parked vehicle is less than a second distance specified.

3. The parking location management device according to claim 2, wherein: The parking place management device includes a distribution processing unit that distributes information on the roadside parking place detected by the parking place detection unit to a vehicle.

4. The parking location management device according to claim 3, wherein: The distribution processing unit is configured as follows: A roadside parking location notification packet is distributed to a vehicle scheduled to pass through the roadside parking location, wherein the roadside parking location notification packet is a communication packet indicating information about the roadside parking location, and When the presence state determination unit determines that the vehicle has disappeared from the roadside parking spot, a disappearance notification packet is distributed to the vehicle to which the roadside parking spot notification packet regarding the roadside parking spot has been distributed. The disappearance notification packet is a communication packet indicating that the parked vehicle has disappeared.

5. The parking location management device according to claim 3, wherein: The parking place management device includes a grouping unit that groups the plurality of roadside parking places detected by the parking place detection unit based on positional relationships among the plurality of roadside parking places detected by the parking place detection unit to thereby set a roadside parking area including the plurality of roadside parking places. The distribution processing unit is configured to distribute the roadside parking points grouped by the grouping unit as roadside parking area information including the starting position coordinates of the roadside parking area.

6. The parking location management device according to claim 5, wherein: The grouping unit is configured to set the roadside parking area by grouping the roadside parking spots whose intervals in the road extending direction are equal to or smaller than a predetermined threshold.

7. The parking location management device according to claim 5, wherein: The distribution processing unit is configured to distribute the roadside parking area information, wherein the roadside parking area information includes, in addition to the position coordinates of the vehicle at the beginning of the roadside parking area, at least one of the position coordinates of the vehicle at the end of the roadside parking area and the length of the roadside parking area.

8. The parking location management device according to any one of claims 5 to 7, wherein: The distribution processing unit is configured to distribute the roadside parking area information, wherein the roadside parking area information includes at least one of characteristic information of a leading vehicle in the roadside parking area and characteristic information of a vehicle located at the end of the roadside parking area. The characteristic information includes at least one of the color of the vehicle body and the type of vehicle.

9. The parking location management device according to claim 3, wherein: The parking spot detection unit is configured to detect the roadside parking spot on the road on the exit side of the intersection with priority over the roadside parking spot on the road on the entrance side of the intersection.

10. The parking location management device according to any one of claims 3 to 7, wherein: The judgment material acquisition unit is configured to acquire the vehicle behavior data from a plurality of vehicles. The parked location detection unit and the continued state determination unit are configured to determine whether an avoidance action of shifting the driving position from the first lane to the second lane is being taken based on the vehicle behavior data. The parking location detection unit determines that a vehicle is parked at a location as a processing target based on the fact that at least one vehicle is performing the avoidance action near the location. The existence state determination unit determines that the vehicle parked at the parking spot has disappeared based on the presence of a vehicle that passes through the parking spot without performing the avoidance action.

11. The parking location management device according to any one of claims 3 to 7, wherein: The judgment material acquisition unit is configured to acquire the sensing information from a plurality of vehicles. The parking location detection unit detects the parking location based on the sensing information provided from a plurality of vehicles. The existence state determination unit determines whether a vehicle still exists at the parking spot based on the sensing information provided from a vehicle passing around the parking spot.

12. The parking location management device according to claim 11, wherein: The parking spot detection unit is configured to determine, based on the sensing information, a lateral length of the first lane blocked by the vehicle parked at the roadside parking spot as a protrusion amount. The distribution processing unit is configured to distribute the protrusion amount in association with the position information of the roadside parking spot.

13. The parking location management device according to claim 1 or 3, wherein: The distribution processing unit sets and distributes a lane change recommendation POI at a location a predetermined distance before the roadside parking location in a lane for parked vehicles.

14. The parking location management device according to claim 1 or 3, wherein: The distribution processing unit sets a section where the number of available lanes is insufficient due to parked vehicles on the road as an automatic driving non-compliant section and distributes the section.

15. A parking location management method, which is a method for managing location information of vehicles parked on a road, executed using at least one processor, wherein: include: a determination material acquisition step of acquiring at least one of vehicle behavior data representing the behavior of at least one vehicle and sensing information from a periphery monitoring sensor mounted on the vehicle in correspondence with the position information; a parking location detection step of detecting a location where a vehicle is parked on a general road, i.e., a roadside parking location, based on the information obtained in the determination material acquisition step; a presence determination step of determining whether there is still a vehicle at the roadside parking location detected in the parking location detection step based on the information obtained in the determination material acquisition step; as well as a distribution step of distributing information about the roadside parking locations to vehicles, The above-mentioned parking location detection steps include: a step of detecting a vehicle stopped on the road, i.e. stopping the vehicle; Based on the fact that the distance between the stopped vehicle and the road end is within a predetermined distance, setting the location where the vehicle is located as the roadside parking location; and A step of detecting the roadside parking spot located on the road on the exit side of the intersection with priority over the roadside parking spot located on the road on the entrance side of the intersection.

16. A parking location management method, which is a method for managing location information of vehicles parked on a road, executed using at least one processor, wherein: include: a determination material acquisition step of acquiring at least one of vehicle behavior data representing the behavior of at least one vehicle and sensing information from a periphery monitoring sensor mounted on the vehicle in correspondence with the position information; a parking location detection step of detecting a location where the vehicle is parked on a general road, i.e., a roadside parking location, based on the information obtained in the determination material acquisition step; and a presence determination step of determining whether there is still a vehicle at the roadside parking location detected in the parking location detection step based on the information obtained in the determination material acquisition step; In the parking location detection step, the location where the parked vehicle is located is detected as the roadside parking location based on the fact that there is no intersection or traffic light within a first distance specified on the road travel direction side from the parked vehicle and the distance from the road end to the parked vehicle is less than a second distance specified.

17. A vehicle device for transmitting information on a location where a vehicle is parked on a road, namely, a parking location, to a predetermined server, wherein: have: a stopped vehicle information acquiring unit that acquires information about stopped vehicles, which are other vehicles stopped on the road, based on an input signal from a surrounding monitoring sensor mounted on the vehicle; a parking determination unit that determines, based on the information about the parked vehicle acquired by the parked vehicle information acquisition unit, whether the parked vehicle corresponds to a roadside parked vehicle or a temporarily parked vehicle that remains movable; as well as a notification processing unit that, when the parked vehicle determined by the parking determination unit to be equivalent to the roadside parked vehicle exists, transmits a data set indicating a location where the roadside parked vehicle exists to the server; The stopped vehicle information acquiring unit acquires the position and direction of the stopped vehicle. When a row of the plurality of parked vehicles includes vehicles whose bodies face in opposite directions to other vehicles, the parking determination unit determines that the parked vehicles forming the row are the roadside parked vehicles.

18. A vehicle device for transmitting information on a location where a vehicle is parked on a road, namely, a parking location, to a predetermined server, wherein: have: a parking determination unit for determining whether the stopped vehicle, which is another vehicle stopped on the road, corresponds to a roadside parked vehicle parked on the road or a temporarily parked vehicle that remains movable; as well as a notification processing unit that, when the parked vehicle determined by the parking determination unit to be equivalent to the roadside parked vehicle exists, transmits a data set indicating a location where the roadside parked vehicle exists to the server; The parking determination unit determines whether there is a parked vehicle on the roadside based on the fact that there is no intersection or traffic light within a predetermined first distance from the stopped vehicle in the road traveling direction and the distance from the road end to the stopped vehicle is less than a predetermined second distance.

19. The vehicle device according to claim 18, wherein The parking determination unit is configured not to determine that the stopped vehicle, the distance of which from the road edge is equal to or greater than a predetermined parking determination threshold, is a roadside parked vehicle.

20. The vehicle device according to claim 17, wherein The vehicle device includes a vehicle behavior acquisition unit that detects a parking state of the vehicle as the behavior of the vehicle. The report processing unit is configured as follows: Based on the input signals from the surrounding monitoring sensors, the surrounding environment of the vehicle when it is parked is determined, and When the host vehicle is parked along the road edge, a data set including the parked position of the host vehicle is transmitted to the server.

21. The vehicle device according to claim 17 or 18, wherein: The vehicle further includes a setting processing unit configured to set a lane change recommended POI.

22. The vehicle device according to claim 17 or 18, wherein: The system further includes a setting processing unit configured to set a section where the number of available lanes is insufficient due to the parked vehicles as a non-automatic driving section.

Images