Electronic control device and parking lot map generation method

Abstract

Provided is an electronic control device that generates a course structure in a parking lot, the electronic control device comprising: an information acquisition unit that acquires other vehicle information regarding a plurality of other vehicles detected in the vicinity of a vehicle by a sensor mounted on said vehicle; a parking section row identification unit that identifies a plurality of partial parking section rows on the basis of information regarding the position and orientation of other stopped vehicles, the information thereof being included in the other vehicle information; a partial course estimation unit that calculates a plurality of reference lines indicating the layout of the parking section rows on the basis of the layout of other vehicles stopped within the partial parking section rows, and estimates a plurality of partial courses by shifting the calculated reference lines by a prescribed distance in a direction perpendicular to the plurality of reference lines; a parking lot map generation unit that generates a course structure in the vicinity of the vehicle by combining the plurality of partial courses that were estimated; and an information output unit that outputs the generated course structure.

Description

Electronic control device and parking lot map generation method Incorporation by reference

[0001] This application claims the priority of Japanese Patent Application No. 2025-6616, which was filed on January 17, 2027 (2025), and incorporates its content by reference into this application.

[0002] The present invention relates to an electronic control device.

[0003] In recent years, technologies have been proposed that construct a map of a parking lot from recognition information of external sensors during parking lot driving and support parking actions in the same parking lot in subsequent times.

[0004] As the background art in this technical field, there are the following prior arts. Patent Document 1 (Japanese Patent Application Laid-Open No. 2020-152234) discloses a sensor input unit that acquires sensor information, which is the output of sensors that acquire information around a vehicle, a movement information acquisition unit that acquires vehicle movement information, which is information related to the movement of the vehicle, a log recording unit that records information based on the vehicle movement information and the sensor information in a storage unit, and a map generation unit that creates an environmental map including a drivable area where the vehicle can drive based on the determination of static objects that do not move and dynamic objects that can move using the information based on the vehicle movement information recorded in the storage unit and the sensor information, and a route calculation unit that calculates the driving route of the vehicle using the environmental map.

[0005] In the invention described in Patent Document 1, by driving through a parking lot once, a map including the drivable area of the parking lot is created, enabling support for parking actions in subsequent times. However, there is a problem that for a parking lot that has never been used before or an area in a parking lot that has been used but has never been driven through, a map cannot be created and parking actions cannot be supported.

[0006] A typical example of the invention disclosed in this application is as follows: an electronic control device for generating a road structure in a parking lot, comprising: an information acquisition unit that acquires information about a plurality of other vehicles detected around the vehicle by a sensor mounted on the vehicle; a parking space row identification unit that identifies a plurality of partial parking space rows based on information about the position and orientation of other vehicles that are stopped, which is included in the other vehicle information; a partial road estimation unit that calculates a plurality of reference lines indicating the arrangement of parking space rows based on the arrangement of other vehicles that are stopped in the partial parking space rows, and estimates a plurality of partial roads by shifting the calculated plurality of reference lines by a predetermined distance in the vertical direction; a parking lot map generation unit that generates a road structure around the vehicle by combining the estimated plurality of partial roads; and an information output unit that outputs the generated road structure.

[0007] According to one aspect of the present invention, it is possible to generate a track structure in a parking area that has never been driven on, thereby supporting parking behavior. Problems, configurations, and effects other than those described above will be clarified by the following description of embodiments.

[0008] This is a functional block diagram showing the configuration of a vehicle system including an electronic control device according to an embodiment of the present invention. This is a diagram showing an example of the installation of a group of external sensors on a vehicle according to this embodiment. This is a diagram showing an example of a parking lot structure for explaining the operation of this embodiment. This is a diagram showing an example of the data structure of the external sensor data group according to this embodiment. This is a diagram showing an example of the data structure of the parking space row data group according to this embodiment. This is a diagram showing an example of the partial road data group according to this embodiment. This is a diagram showing an example of the parking lot map data group according to this embodiment. This is a functional block diagram of the electronic control device according to this embodiment. This is a flowchart of the processing performed by the parking space row identification unit according to this embodiment. This is a diagram showing an example of the processing in steps S902 to S9034 to S906 according to this embodiment. This is a diagram showing an example of the processing in steps S907 to S910 according to this embodiment. This is a diagram showing an example of applying the clustering method of this embodiment to recognition information of parking spaces arranged diagonally. This is a diagram showing an example of the processing performed by the partial road estimation unit according to this embodiment. This shows an example of the processing in steps S1402 to S1406 according to this embodiment. This shows an example of the processing in steps S1402 to S1406 according to this embodiment. This is a diagram showing an example of the process performed by the parking map generation unit of this embodiment. This is a diagram showing an example of the process performed by the parking map generation unit of this embodiment. This is a diagram showing an example of the process performed by the parking map generation unit of this embodiment. This is a flowchart of the process performed by the driving plan unit of this embodiment. This is a diagram showing an example of the screen display information of this embodiment.

[0009] <Embodiment 1> Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0010] (System Configuration) Figure 1 is a functional block diagram showing the configuration of a vehicle system 1 including an electronic control device 3 according to an embodiment of the present invention.

[0011] Vehicle system 1 is mounted on vehicle 2. Vehicle system 1 understands the situation of obstacles such as parking spaces and surrounding vehicles around vehicle 2 and provides appropriate driving assistance and driving control, especially for parking in parking lots. As shown in Figure 1, vehicle system 1 consists of an electronic control unit 3, an external sensor group 4, a vehicle sensor group 5, an actuator group 7, and an HMI device group 8. The electronic control unit 3, external sensor group 4, vehicle sensor group 5, actuator group 7, and HMI device group 8 are connected by an in-vehicle network N. Hereafter, vehicle 2 may be referred to as "our vehicle" 2 to distinguish it from other vehicles.

[0012] The electronic control unit 3 is an ECU (Electronic Control Unit). Based on various input information provided by the external sensor group 4, the vehicle sensor group 5, etc., the electronic control unit 3 generates driving plan information for parking assistance of the vehicle 2 and outputs it to the actuator group 7 and the HMI device group 8. The electronic control unit 3 has a processing unit 10, a storage unit 30, and a communication unit 40.

[0013] The processing unit 10 is configured to include, for example, a central processing unit, which is a CPU (Central Processing Unit). However, it may also be configured to include a GPU (Graphics Processing Unit), FPGA (Field-Programmable Gate Array), ASIC (Application Specific Integrated Circuit), etc., in addition to the CPU, or it may be configured to include any one of them.

[0014] The processing unit 10 has the following functions: an information acquisition unit 11, a parking space row identification unit 12, a partial road estimation unit 13, a parking lot map generation unit 14, a driving plan unit 15, and an information output unit 16. The processing unit 10 achieves these functions by executing a predetermined operation program stored in the storage unit 30.

[0015] The information acquisition unit 11 acquires various information from other devices connected to the electronic control unit 3 via the in-vehicle network N and stores it in the storage unit 30. For example, it acquires external sensor data group 31, which includes information about stationary and moving objects around the vehicle 2 detected by the external sensor group 4, and vehicle information data group 32, which relates to the movement and state of the vehicle 2 detected by the vehicle sensor group 5, etc., and stores them in the storage unit 30.

[0016] The parking space row identification unit 12 identifies the parking space row around the vehicle 2 based on recognition information regarding parked vehicles and parking spaces included in the external sensor data group 31 acquired by the information acquisition unit 11 and stored in the storage unit 30, and stores it in the storage unit 30 as a parking space row data group 33. A parking space row is a collection of parking spaces arranged in a continuous line in the same direction.

[0017] The partial road estimation unit 13 estimates the partial roads adjacent to a parking space based on the arrangement of parking spaces in the parking space row data group 33, and stores them in the storage unit 30 as partial road data group 34.

[0018] The parking lot map generation unit 14 integrates the partial road data group 34 in chronological order to construct the road structure of the parking lot and stores it in the storage unit 30 as a parking lot map data group 35.

[0019] The driving plan unit 15 determines whether the conditions for automatic driving of the vehicle 2 are met based on the vehicle information data group 32 and the parking map data group 35, etc. If automatic driving is possible, it generates driving plan information for automatic driving and stores it in the storage unit 30 as driving plan data group 36. Then, in manual driving mode, the driving plan unit 15 generates information to present the recommended route for automatic driving, etc., to the occupant via the HMI device group 8, and in automatic driving mode, it generates information for controlling the actuator group 7 in addition to the above.

[0020] The information output unit 16 outputs various information to other devices connected to the electronic control unit 3 via the in-vehicle network N. For example, the information output unit 16 outputs control information included in the driving plan data group 36 determined by the driving plan unit 15 to the actuator group 7 to control the driving of the vehicle 2. Also, for example, the information output unit 16 outputs recommended route information included in the driving plan data group 36 to the HMI device group 8 to support the occupant's parking actions and enable the occupant to understand the vehicle 2's driving during autonomous driving.

[0021] The storage unit 30 is comprised of, for example, storage devices such as HDDs (Hard Disk Drives), flash memory, and ROMs (Read Only Memory), as well as non-volatile storage media such as RAMs. The storage unit 30 stores programs processed by the processing unit 10 and data sets necessary for that processing. It may also be used as the main memory when the processing unit 10 executes a program, temporarily storing data necessary for program calculations. External sensor data groups 31, vehicle information data groups 32, parking space row data groups 33, partial road data groups 34, parking map data groups 35, driving plan data groups 36, etc., are stored in the storage unit 30.

[0022] The external sensor data group 31 is a collection of data relating to recognition information recognized by the external sensor group 4. Recognition information includes, for example, information about environmental elements such as other vehicles or parking spaces that the external sensor group 4 identifies based on its sensing information.

[0023] The vehicle information data group 32 is a collection of data relating to the movement and state of the vehicle 2, and includes vehicle information detected by the vehicle sensor group 5 and acquired by the information acquisition unit 11. The vehicle information includes, for example, information such as the position of the vehicle 2, driving speed, steering angle, accelerator operation amount, brake operation amount, and driving mode.

[0024] The parking space row data group 33 is a collection of data relating to the parking space row identified by the parking space row identification unit 12.

[0025] The partial track data set 34 is a collection of data relating to partial tracks identified by the partial track estimation unit 13.

[0026] The parking lot map data set 35 is a collection of data related to parking lot maps generated by the parking lot map generation unit 14, which represent the structure of the parking lot's roads, direction of travel, etc.

[0027] The driving plan data group 36 is a collection of data related to recommended driving route information generated by the driving plan unit 15 and driving control plan information for driving the recommended driving route.

[0028] The communication unit 40 is configured to include, for example, a network card compliant with communication standards such as IEEE 802.3 or CAN (Controller Area Network), and transmits and receives data with other devices of the vehicle system 1 according to various protocols.

[0029] In this embodiment, the communication unit 40 and the processing unit 10 are described separately, but some of the processing of the communication unit 40 may be executed within the processing unit 10. For example, hardware devices equivalent to those in communication processing may be provided in the communication unit 40, while other device drivers and communication protocol processing may be provided in the processing unit 10.

[0030] The external sensor group 4 is a collection of devices capable of detecting the conditions around the vehicle 2. Examples of external sensor group 4 include cameras, millimeter-wave radar, LiDAR, sonar, etc. The external sensor group 4 detects environmental elements such as other vehicles and parking spaces within a predetermined range from the vehicle 2 and outputs them to the in-vehicle network N. It is preferable for the external sensor group 4 to identify detection targets using not only single observation data but also past time-series observation data and detection results. Alternatively, the detection targets may be identified by integrating observation data from multiple sensor devices of the external sensor group 4.

[0031] The vehicle sensor group 5 is a collection of devices that detect various states of the vehicle 2. Each vehicle sensor detects, for example, the vehicle 2's position information, driving speed, steering angle, accelerator operation amount, brake operation amount, etc., and outputs them to the in-vehicle network N.

[0032] The actuator group 7 is a group of devices that control control elements such as steering, brakes, and accelerators that determine the movement of the vehicle. The actuator group 7 controls the movement of the vehicle based on operation information from the occupants using the steering wheel, brake pedal, accelerator pedal, etc., and control information output from the electronic control unit 3.

[0033] The HMI device group 8 is a group of devices for receiving information input from the occupant to the vehicle system 1 and for notifying the occupant of information from the vehicle system 1. The HMI device group 8 includes a display, speaker, vibrator, switch, etc.

[0034] Figure 2 shows an example of the installation of the external sensor group 4 on the vehicle 2.

[0035] In the example shown in Figure 2, regions 111 to 115 are formed that can be detected by the external sensor group 4, and these regions enable the detection of environmental elements around the entire vehicle 2. The external sensor group 4 may be implemented with a single sensor device, such as LiDAR, or with a combination of multiple sensor devices, such as a multi-camera. In this embodiment, it is not necessarily required to be able to recognize the entire surroundings; it is sufficient if it is able to recognize a predetermined range around the vehicle 2.

[0036] Figure 3 shows an example of a parking lot structure for explaining the operation of this embodiment.

[0037] The parking lot shown in Figure 3 is structured so that vehicles enter from the entrance 301 and exit from the exit 302, and can travel along the dashed road 305. Arrows 306 indicate the directions in which travel is permitted on the road 305. For example, arrow 306 means that vehicles can only proceed in the left-turn direction from the entrance. In the parking lot, parking spaces (e.g., 307), which are spaces in which vehicles can park, are arranged in a line according to predetermined rules. In this embodiment, a group of parking spaces that are lined up continuously in the same direction is called a "row of parking spaces". In the parking lot shown in Figure 3, there are rows of parking spaces 311 to 315. Even if parking spaces are lined up together in the same place, such as rows 311 and 312, if they are lined up in two rows, each is considered a separate row of parking spaces. Alternatively, all parking spaces included in a row of parking spaces face the same road (it is assumed that vehicles enter from the same road). For example, parking space row 311 faces the road shown above, and parking space row 312 faces the road shown below. Since parking space row 311 and parking space row 312 are accessible from different roads, they are treated as separate parking space rows.

[0038] Figure 4 shows an example of the data structure of the external sensor data group 31 in this embodiment.

[0039] The external sensor data group 31 consists of time 401, ID 402, type 403, position and orientation 404, width and length 405, speed 406, and confidence level 407, etc.

[0040] Time 401 is information indicating the time of the data entry. This may be the time the data entry was observed by the external sensor group 4, or the time the electronic control unit 3 received the data.

[0041] ID 402 is identification information used to determine whether data entries acquired at different times are targeting the same environmental element. Recognition information for the same environmental element will store the same identification information. Information detected by different external sensors may be assigned different identification information even if it is the same object, or recognition information for the same object may be identified through sensor fusion, etc., and the data entries may be integrated to assign a single identification information.

[0042] Category 403 is the identification result of the detected object. For example, information such as "other vehicle" and "parking space" is stored.

[0043] The position and orientation 404 is information regarding the relative position and direction seen from the vehicle 2 with respect to the representative point and orientation of the detected object. For example, when the object is a vehicle, the representative point is the center point of a figure representing the outline of the other vehicle as a rectangle (two-dimensional view from above) or a rectangular parallelepiped (three-dimensional), and the representative orientation is the forward direction of the other vehicle. When the object is a parking space, the representative point is the center point of the parking space, and the representative orientation is the forward direction of the vehicle to be parked. Note that the orientation of the other vehicle or the parking space may be reversed depending on the recognition status of the external sensor group 4, but as will be described later, the process for handling this data can execute the process without problems even if the orientation is reversed. The relative position and direction seen from the vehicle 2 means the position and orientation of the object when expressed in a coordinate system with the center of the rear wheel axle of the vehicle 2 as the origin, the forward direction of the vehicle 2 as the positive x-axis direction, and the leftward direction of the vehicle 2 as the positive y-axis direction.

[0044] The width / length 405 is information regarding the length "width" of the side perpendicular to the representative orientation and the length "length" of the side parallel thereto when the outline of the detected object is represented as a rectangle (two-dimensional view from above).

[0045] The speed 406 is information regarding the relative speed of the object seen from the vehicle 2. For example, the ground speed vector of the object is separated and expressed as the x-axis component and y-axis component of the vehicle 2.

[0046] The confidence level 407 is a quantification of the degree of certainty of the information of the object detected by the external sensor group 4. For example, it is expressed as a numerical value from 0.0 to 1.0, and the larger the numerical value, the more certain the recognition of the object.

[0047] FIG. 5 is a diagram showing an example of the data structure of the parking section column data group 33 of the present embodiment.

[0048] The parking section column data group 33 is configured to include a time 501, a section column ID 502, a parking section ID 503, a position and orientation 504, a width / length 505, a speed 506, a confidence level 507, and the like.

[0049] Each data entry is configured as information for each parking space. The information for each parking space is generated based on the parked vehicle and parking space information from the external sensor data group 31. The time 501, position and orientation 504, width and length 505, and confidence level 507 are the same as the time 401, position and orientation 404, width and length 405, and confidence level 407 of the external sensor data group 31 shown in Figure 4, respectively. However, each data in the parking space column data group 33 is a value determined by considering past time-series data and may differ from the sensor output values ​​of the external sensor data group 31. The space column ID 502 is the identification information of the parking space column to which the parking space of the data entry is classified. The parking space ID 503 is the identification information of the parking space included in the parking space column and is generated based on the ID 402 of the external sensor data group 31.

[0050] Figure 6 shows an example of a partial road data group 34 in this embodiment. The partial road data group 34 is a collection of information about partial roads estimated from the parking space row data group 33.

[0051] In Figure 6, partial lanes 601 to 603 are shown between parking space rows 311 to 315. Note that the information regarding parking space rows 311 to 315 is shown for explanatory purposes to clarify the positional relationship with the parking example in Figure 3, and is not included in the actual partial lane data set 34.

[0052] A section of the track is represented by a reference line of the track, and is represented according to a relative coordinate system as seen from vehicle 2. The reference line of the track is a sequence of points that represents the shape of the track, preferably the shape of the track's centerline. However, in this embodiment, the reference line of the track is mainly used to calculate a recommended route for vehicle 2 to travel within the parking lot, so it is not a problem if the position of the reference line is slightly off, and the purpose is to represent the shape of the section of the track and the relationship between the section of the track.

[0053] For example, the partial track 601 shown in Figure 6 is represented by two shape points 611 and 612. The line connecting these adjacent shape points (generally a polyline) is the baseline of the track. While a straight partial track like the one in Figure 601 can be represented by two points at both ends, if the partial track is curved, additional shape interpolation points are added to represent the curved shape, in addition to the two points at both ends.

[0054] Figure 7 shows an example of the parking lot map data set 35 in this embodiment.

[0055] The parking lot map data set 35 is a collection of information obtained by accumulating and integrating the partial road data set 34 in chronological order to estimate the road structure of the parking lot.

[0056] The parking lot map data set 35 should, for example, be structured in accordance with the structure of general digital map data. Specifically, it consists of "links" that indicate partial roads and "nodes" that are the connection points between links.

[0057] Each link contains information such as identification information to uniquely identify the link, a sequence of shape points representing the path shape formed by the link, a path direction indicating the drivable direction of travel, identification information for the nodes at both ends, and identification information for the associated parking space sequence. For example, the parking lot map shown in Figure 7 is represented by links 701 to 707 and nodes 711 to 719. In addition, information on path directions 722 to 724 indicating the drivable direction of travel is attached to links 702 to 704. For links where a path direction is not indicated, the path direction is unknown at this point. Note that the information regarding parking space sequences 311 to 315 is shown for explanatory purposes to clarify the positional relationship with the example parking lot in Figure 3, and the actual parking lot map data set 35 does not include information regarding parking space sequences 311 to 315.

[0058] For example, link 704 includes information such as identification information 704, shape point sequence which is the position coordinates of nodes 716 and 713, travel direction 724 which is the direction from node 716 to node 713, identification information for both end nodes 716 and 713, and identification information for the associated parking space sequence which is the sequence ID of parking space sequences 312 and 313 (502 in Figure 5). The shape point sequence is synonymous with the baseline of the partial road, and if it is a straight line, it is represented by the position coordinates of two points of the nodes at both ends, but if it is a curve, it must include the position coordinates of shape interpolation points sufficient to represent the shape of the link. The associated parking space sequence is the parking space sequence facing the partial road represented by the link. The parking space sequence information allows for the identification of parking spaces that can be accessed from the link.

[0059] Each node contains information such as identification information to uniquely identify the node, the node's position coordinates, and a list of identification information for the links connected to the node. For example, the identification information for the links connected to node 716 would be 704, 705, and 706. By tracing the identification information of the nodes included in the links and the identification information of the links included in the nodes, the connection relationships of the sub-paths can be calculated.

[0060] The operation of the vehicle system 1 will be explained with reference to Figures 8 to 15.

[0061] Figure 8 is a functional block diagram of the electronic control device 3 of this embodiment.

[0062] The electronic control unit 3 is configured such that the processing of the information acquisition unit 11, parking space row identification unit 12, partial road estimation unit 13, parking map generation unit 14, driving plan unit 15, and information output unit 16 in Figure 1 is executed in an appropriate order. The series of processes is repeatedly executed at predetermined time intervals (for example, every 100 ms).

[0063] The information acquisition unit 11 acquires necessary information from other devices via the in-vehicle network N and stores the acquired information in the storage unit 30. For example, it acquires the external sensor data group 31 from the external sensor group 4 and the vehicle information data group 32 from the vehicle sensor group 5, and passes them on to the subsequent processing unit.

[0064] The parking space row identification unit 12 extracts elements of a parking space based on recognition information regarding parked vehicles and parking spaces around the vehicle 2 included in the external sensor data group 31, identifies the parking space row around the vehicle 2 based on the extracted parking space elements, and outputs the identified parking space row as a parking space row data group 33.

[0065] The partial road estimation unit 13 estimates the partial roads adjacent to the parking spaces based on the arrangement of the parking spaces in the parking space row data group 33, and outputs the estimated partial roads as a partial road data group 34.

[0066] The parking lot map generation unit 14 generates and outputs a parking lot map data group 35 based on the parking space row data group 33 and the partial road data group 34. It is preferable to obtain the parking lot map data group 35 generated in a previously executed process from the storage unit 30, generate data corrected for the elapsed time using the vehicle information data group 32, and then integrate the parking space row data group 33 and partial road data group 34 generated based on the new recognition information external sensor data group 31 with the corrected parking lot map data group 35 to generate the parking lot map data group 35.

[0067] The driving plan unit 15 determines whether the vehicle 2 meets the conditions for autonomous driving based on the vehicle information data group 32, the parking map data group 35, etc., and if autonomous driving is possible, it generates driving plan information for autonomous driving and outputs the generated driving plan information as driving plan data group 36.

[0068] The information output unit 16 outputs various information to other devices connected to the electronic control unit 3 via the in-vehicle network N. For example, the information output unit 16 outputs control information included in the driving plan data group 36 determined by the driving plan unit 15 to the actuator group 7 to control the driving of the vehicle 2. Also, for example, the information output unit 16 outputs recommended route information and the like included in the driving plan data group 36 to the HMI device group 8 to present to the occupants.

[0069] (Processing by the parking space row identification unit 12) The process performed by the parking space row identification unit 12 will be explained using the flowchart in Figure 9.

[0070] First, in step S901, the parking space row identification unit 12 acquires the external sensor data group 31.

[0071] Next, in step S902, the parking space row identification unit 12 extracts recognition information regarding parked vehicles and parking spaces from the acquired external sensor data group 31. First, it refers to the type 403 of the external sensor data group 31 (Figure 4) to extract data entries related to other vehicles and parking spaces. The data entries for other vehicles include not only parked vehicles but also moving vehicles on the road. To extract parked vehicles from these, for example, vehicles with a speed 406 close to 0 are selected. Moving vehicles may stop temporarily, making it difficult to determine their status based on information at a single point in time in the data entry. Therefore, the same other vehicle may be tracked using ID 402, and it may be determined whether it is moving based on changes in position or speed over time. At this point, it is not necessary to completely distinguish between parked vehicles and moving vehicles, and moving vehicles may be included.

[0072] In step S903, the extracted data entries for parked vehicles and parking spaces are normalized into a common data structure called parking space information PSR(t). Both parked vehicles and parking spaces are information that suggests the location of a parking space. In subsequent processing, in order to estimate the parking space sequence from the position and orientation of the parking spaces, the information of parked vehicles and parking spaces, which have different data structures, is converted into a common data structure called a parking space to facilitate processing. The data structure of a parking space can be a single data entry as shown in Figure 5.

[0073] In step S904, the parking space row information PSL(t-1) from the previous processing is obtained from the parking space row data group 33. Then, in step S905, the parking space row information PSL(t-1) from the previous processing is corrected to reflect the positional relationship at the target time t of the current processing. If vehicle 2 has moved between time t-1 and time t, the positional relationship of the parking space row as seen from vehicle 2 changes. The amount of movement of vehicle 2 is estimated using the position information and speed information of vehicle 2 included in the vehicle information data group 32, and the information of each parking space in the parking space row information PSL(t-1) is corrected using the estimated amount of movement.

[0074] In step S906, the parking space information PSR(t) normalized in step S903 and the parking space information contained in PSL(t-1) corrected in step S905 are integrated. In the integration process of step S906, time-series recognition information tracking technology can be used. The relationship between each parking space in the corrected PSL(t-1) and each parking space in the recognition information PSR(t) is determined, and information regarding the same parking space is integrated. If only PSR(t) exists, it is added as new parking space information.

[0075] In step S907, the integrated parking space information is primary clustered on a parking space basis. The parking space row identification unit 12 performs clustering of the parking space information in two stages. The first clustering is called primary clustering, and the second clustering is called secondary clustering. In primary clustering, the parking spaces are roughly clustered based on their positional relationship and orientation. Then, in step S908, the orientation of the parking spaces is estimated for each cluster of roughly clustered parking spaces, and in step S909, the parking spaces in each cluster are further secondary clustered using the estimated orientation. Details of each clustering process will be described later with reference to Figure 12.

[0076] In step S910, among the clusters for which secondary clustering has been completed, clusters containing a predetermined number or more parking spaces are stored in the parking space column data group 33 as new parking space column data PSL(t). Clusters with a small number of parking spaces do not need to be treated as parking space columns, as this means that only a small number of parking spaces are located close together and facing similar directions. Note that moving vehicles that could not be excluded in step S902 will be in a different position and orientation from the arrangement of parking space columns, and can therefore be excluded in step S910.

[0077] The process performed by the parking space row identification unit 12 will be explained with reference to Figures 10A to 13.

[0078] Figures 10A and 10B show examples of the processes in steps S902 and S903.

[0079] Figure 10A shows the recognition information included in the external sensor data group 31 at time t of vehicle 2.

[0080] The recognition information shown in Figure 10A includes recognition information for parking spaces 1011 and 1012, and recognition information for other vehicles 1021, 1031, and 1032. Other vehicles 1031 and 1032 are moving vehicles. In step S902, the information shown in Figure 10A is basically extracted. Moving vehicles 1031 and 1032 are preferably excluded from the extraction target in step S902, however, moving vehicle 1031 may be treated as a parked vehicle because it waits for moving vehicle 1032 to pass and remains stopped for a predetermined time.

[0081] Figure 10B shows the parking space information PSR(t) generated in step S903. The recognition information shown in Figure 10A has been converted into the format of parking space information, where each rectangle represents the position and orientation 404, width and length 405 in Figure 4, and the number inside each rectangle represents the confidence level 407. In Figure 10A, parking spaces 1011 and 1012 also have recognition information of other vehicles overlapping, but information is integrated so that those that are considered to be in the same parking space area become a single parking space information, and parking space information is generated.

[0082] Figure 11 shows an example of the process in steps S904 to S906.

[0083] Figure 11(A) shows the parking space row information PSL(t-1) acquired in step S904 after the position correction in step S905. At time t-1, vehicle 2 is at position 1110, so PSL(t-1) is expressed in coordinates relative to position 1110. In order to convert the previously processed recognition information into coordinates relative to the position of vehicle 2 at time t, the amount of movement of vehicle 2 between time t-1 and time t is estimated from the vehicle information data group 32, etc., and the recognition information is corrected using the estimated amount of movement.

[0084] Figure 11(B) shows the parking space information PSR(t) from Figure 10B.

[0085] Figure 11(C) shows the integrated data of the parking space information PSL(t-1) (Figure 11(A)) and parking space information PSR(t) (Figure 11(B)) after position correction. Based on the positional relationship of each parking space in both sets of data, the correspondence between the parking spaces is determined.

[0086] For identical parking spaces, the corrected parking space information PSL(t-1) is updated based on the parking space information PSR(t). For example, if the confidence level of parking space information PSR(t) is higher than that of parking space information PSL(t-1), the position, orientation, width / length, confidence level, etc., may be overwritten with the parking space information PSR(t). Alternatively, the parking space information PSL(t-1) and parking space information PSR(t) may be probabilistically integrated using Bayesian estimation or a Kalman filter. Although the confidence level does not directly represent the probability of a parking space's existence, it is possible to convert it into information equivalent to the probability based on its calculation method and handle it accordingly.

[0087] Among the parking spaces (e.g., 1121, 1122) that exist only in the parking space row information PSL(t-1) in Figure 11(A), those with a confidence level of a predetermined value or higher may be retained, while those with a confidence level lower than the predetermined value may be deleted. Alternatively, when integrating probabilistically, the probability may be lowered based on the fact that the recognition information 1002 has not detected the parking space, and parking spaces whose probability falls below a predetermined value may be deleted.

[0088] The parking spaces that exist only in the parking space information PSR(t) in Figure 11(B) (for example, 1111-1114) are newly detected parking spaces and are added to the parking space information as new parking space entries.

[0089] Figure 12 shows an example of the process in steps S907 to S910.

[0090] Figure 12(A) shows the results of the primary clustering performed in step S907 using the integrated parking space information from Figure 11(C). In the primary clustering in step S907, clustering is performed on parking spaces with a confidence level of a predetermined value or higher. In Figure 12(A), parking spaces with a confidence level of 0.5 or higher are targeted, and parking spaces 1111 to 1113 are excluded from the clustering process. The reason for targeting parking spaces with high confidence levels in primary clustering is to avoid affecting the estimation of the parking space sequence if spaces with low recognition accuracy are included. Primary clustering is performed when both of the following two conditions are met, for example, (1) the angles of orientation of the parking spaces are close, and (2) the distance between parking spaces is close. In this case, as shown in clusters 1211 and 1212 in Figure 12(A), clusters are set up to include parking spaces that have a common orientation (e.g., within a predetermined angle range) and are relatively close in distance (e.g., the center position of the parking space is within a predetermined distance range).

[0091] Figure 12(B) shows the estimated orientation of the parking space arrangement in step S908, namely, the orientation 1221 of the parking space arrangement in cluster 1211 and the orientation 1222 of the parking space arrangement in cluster 1212. Note that orientations 1221 and 1222 are information similar to direction vectors, and the starting position of the arrows in the figure has no particular meaning. The orientation of the parking space arrangement can be calculated, for example, by analyzing the relative position vectors of the parking spaces and selecting the orientation with the most common vector orientations.

[0092] Figure 12(C) shows the results of the secondary clustering process in S909. Cluster 1211 is re-clustered and divided into clusters 1211-1 to 1211-4. Cluster 1212 is re-clustered and divided into clusters 1212-1 and 1212-2. In secondary clustering, clustering is performed when the condition that (3) the positional deviation in the direction perpendicular to the orientation of the parking space arrangement is close is met. The positional deviation is the signed length of the perpendicular line drawn from the position coordinate of a parking space to a straight line placed at an arbitrary position along the orientation of the parking space arrangement (the sign is ± depending on the direction in which the perpendicular line is drawn). For example, the perpendicular line 1231 of the parking space in cluster 1211-1 and the perpendicular line 1232 of the parking space in cluster 1211-2 are shown. If parking spaces are in the same row along the direction of arrangement, their deviations will be close, but if they are in different rows, there will be a difference in the deviation values. Therefore, by clustering under condition (3), different rows along the direction of arrangement can be separated. This allows sets of parking spaces located in the same "island," such as parking space rows 311 and 312 in Figure 3, to be separated by parking space row.

[0093] Furthermore, in the example of secondary clustering of cluster 1212, information corresponding to moving vehicles that are outside the order of parking spaces can be separated into cluster 1212-2. In this way, moving vehicles mixed in with the original parking spaces can be separated, and the order of parking spaces can be robustly identified. Note that since the separated cluster 1212-2 contains only one piece of information corresponding to a moving vehicle, it is excluded from the target of the parking space order in step S910.

[0094] Furthermore, this clustering method is also effective in parking lots where parking spaces are arranged vertically (i.e., along the longer side of the parking space) or diagonally. Figure 13 shows an example of applying this clustering method to the recognition information of parking spaces arranged diagonally. Even when parking spaces are arranged diagonally, the relationship between the deviation of the parking space's position and the direction of arrangement of the parking spaces remains the same as when the parking spaces are arranged along the shorter side. Therefore, as shown in Figure 13, the parking spaces are ultimately divided into clusters represented by clusters 1301-1, 1301-2, and 1302-1.

[0095] (Processing by the partial track estimation unit 13) The processing performed by the partial track estimation unit 13 will be explained with reference to Figure 14.

[0096] First, in S1401, the partial road estimation unit 13 acquires the parking space row data PSL(t) identified by the parking space row identification unit 12.

[0097] Next, in S1402 to S1406, the partial road estimation unit 13 repeatedly executes the process of generating the base portion of the partial road for each unprocessed row of parking spaces included in PSL(t).

[0098] In S1403, the partial road estimation unit 13 determines a reference line for the unprocessed row of parking spaces psl and generates a shape line S by shifting the determined reference line perpendicularly and toward the vehicle by a predetermined distance d. The reference line is a line that represents the shape of the arrangement of the row of parking spaces, and is, for example, a sequence of points information that connects the reference points (position and orientation 504 in Figure 5) of the parking spaces in the row of parking spaces in the order they are arranged. Alternatively, it may be a line segment that approximates the sequence of points information. Since a road is required to enter the parking space of the row of parking spaces, the road usually exists parallel to the arrangement of the row of parking spaces. Therefore, it is assumed that the road exists at a position shifted parallel to the reference line of the row of parking spaces by a predetermined distance d. The distance d may be set as a fixed value in advance, or the width of the road may be estimated based on information recognized by the external sensor group 4 when the vehicle 2 is driving through the parking lot, and the distance d may be calculated dynamically.

[0099] In S1404, the partial road estimation unit 13 determines whether the shape line S generated in S1403 overlaps with another row of parking spaces or is within a predetermined distance. If the shape line S overlaps with another row of parking spaces or is within a predetermined distance from another row of parking spaces (Y in S1404), the partial road estimation unit 13 discards the shape line S generated in S1403 and generates a shape line S shifted by a predetermined distance d to the opposite side of the vehicle (S1405). On the other hand, if the shape line S does not overlap with another row of parking spaces and is not within a predetermined distance from another row of parking spaces (N in S1404), the partial road estimation unit 13 retains the shape line S generated in S1403. If the shape line S generated on the vehicle's side overlaps with another row of parking spaces, or is located within a predetermined distance of it, it means that there is another row of parking spaces on the vehicle's side of the row of parking spaces psl, and therefore it can be inferred that there is a road on the opposite side from the vehicle.

[0100] In step S1406, the partial track estimation unit 13 generates a partial track R based on the shape lines S generated in S1403 to S1405.

[0101] Figure 15A shows an example of the processing result of steps S1402 to S1406 when the parking space row data PSL(t) shown in Figure 12(C) is acquired. Figure 15A shows the reference lines 1501 to 1505 of clusters 1211-1 to 1211-4 and 1212-1 in Figure 12(C). Since the shape lines 1512, 1513, and 1515 on the vehicle side (the star in the figure indicates the position of vehicle 2) do not overlap with other parking space rows, the reference lines 1502, 1503, and 1505 are used as is (N in S1404). On the other hand, when creating the shape lines on the vehicle side from the reference lines 1501 and 1504, the shape lines created from the parking space rows of cluster 1211-2 (reference line 1502) and reference line 1501, and the shape lines created from the parking space rows of cluster 1211-3 (reference line 1503) and reference line 1504 overlap, respectively. Therefore, the shape lines 1511 and 1514 on the opposite side of the vehicle are adopted (Y in S1404).

[0102] Let's return to the flowchart in Figure 14 and continue the explanation. In steps S1402 to S1406, once processing is complete for all rows of parking spaces (N in S1402), the process proceeds to step S1407. In steps S1407 to S1413, the partial road estimation unit 13 performs processing to supplement the partial road information based on the base portion of the partial road generated in steps S1402 to S1406.

[0103] First, in step S1407, the partial road estimation unit 13 detects overlapping partial roads and merges them into a single partial road. In Figure 15A, shape lines 1512 and 1513 correspond to the same road and are overlapping. For example, the overlap of two shape lines is detected by determining whether the orientation of the two partial roads is approximately the same and the distance between them is close (for example, within a predetermined error range). In merging partial roads, for example, endpoints that are far from the center of each partial road and make the partial road longer are selected. This is because if a parking space exists on at least one of the two corresponding rows of parking spaces, the road must continue to that location. The position of the shape line may be determined, for example, by the average of the positions of both shape lines. Figure 15B shows the shape line 1516 formed by merging shape lines 1512 and 1513.

[0104] Next, in step S1408, the partial road estimation unit 13 extends both ends of each partial road and calculates the intersections between the partial roads. At this stage, the partial roads are estimated only within the range facing the recognized rows of parking spaces, so as shown in Figure 15A, the partial roads are not connected to each other. In order to estimate the connection relationships between the partial roads, the partial roads are extended to find locations (intersections) where they may be connected to each other.

[0105] In step S1409, the partial track estimation unit 13 determines whether the intersection of the partial tracks is valid as a connection, and if so, extends the partial track to the intersection. The validity of the connection is determined, for example, by whether the intersection falls within the assumed area of ​​the parking lot. If the assumed area of ​​the parking lot is unknown, it can be determined, for example, by the distance between the calculated intersection and the endpoint of the partial track. If the intersection is sufficiently close to the endpoint of the partial track, there is a high probability that the two partial tracks are actually connected. On the other hand, if the intersection is far from the endpoint of the partial track, there is a high probability that the intersection was generated due to an error in the shape line even though they are actually parallel and there is no connection point, so the partial track is not extended.

[0106] The partial roads 1511 to 1515 shown in Figure 15A are processed in steps S1407 to S1409 to generate the supplemented partial roads 1521 to 1525 shown in Figure 15B. The base portions 1511, 1516, 1514, and 1515 of the partial roads become partial roads 1521, 1526, 1524, and 1525, respectively. At this point, the information for the partial roads may be structured in a node-and-link data format, similar to parking lot map data. In that case, 1521, 1526, 1524, and 1525 are structured as links, and 1531, 1532, and 1533 are structured as nodes.

[0107] In steps S1410 to S1413, the partial road estimation unit 13 estimates the drivable direction of each partial road using information for moving vehicles and the vehicle itself. The partial roads estimated from the row of parking spaces are determined by their static shape, and it is unknown which direction is drivable. If a moving vehicle is moving on that partial road, it can be determined that it is drivable in the direction the moving vehicle is moving. Therefore, the partial road estimation unit 13 acquires information on the moving vehicle (S1410), identifies the partial road on which the moving vehicle is traveling based on its position, speed, etc. (S1411), and sets the direction of movement of the moving vehicle to the drivable direction of the partial road (S1412). Similarly, it sets the drivable direction of the partial road based on the direction of movement in which the vehicle is (manually) traveling (S1413).

[0108] As described above, by utilizing information about other vehicles and parking spaces around vehicle 2, the road structure around vehicle 2 can be estimated even before vehicle 2 actually starts driving.

[0109] (Processing of Parking Map Generation Unit 14) Figures 16A, 16B, and 16C show examples of the process by which the Parking Map Generation Unit 14 dynamically constructs a parking map by integrating the partial roads estimated by the partial road estimation unit 13 in a time-series manner. In Figures 16A, 16B, and 16C, the colored parking spaces are parking spaces that the Parking Space Column Identification Unit 12 holds as a parking space column data group 33 based on the recognition information up to that point. The links and nodes written around the parking space columns represent the parking map data group 35 constructed by integrating the partial road data group 34 estimated by the partial road estimation unit 13 in a time-series manner. The base information of the partial roads before supplementation by the partial road estimation unit 13 is shown with solid lines, and the supplemented and added partial road information is shown with dashed lines. Figures 16A, 16B, and 16C show how the parking map data group 35 is constructed in a time-series manner while the vehicle 2 is moving.

[0110] In the flowchart of Figure 14, the partial road estimation unit 13 estimates the partial road from the parking space row data, but the partial road may also be estimated by combining it with conventional techniques. For example, by combining it with a technique for estimating the road ahead from an image, it is possible to estimate the road ahead of the vehicle 2 within a predetermined range. In Figures 16A, 16B, and 16C, it is shown that the road ahead of the vehicle 2 within a predetermined range can be estimated.

[0111] Figure 16A shows the state of vehicle 2 when it enters the parking lot. Since it has only recognized some of the vehicles in the row of parking spaces 314, a partial track (link) 1611 is constructed. Subsequently, when vehicle 2 moves to the position shown in Figure 16B, the recognition range of the parking spaces by vehicle 2 expands, and a partial track 1615 is constructed from the recognition information of the rows of parking spaces 312 and 313. Furthermore, a node 1632 is generated from the intersection of the partial tracks 1613 and 1614 that vehicle 2 is traveling on and the extension of partial track 1615, and the partial track that vehicle 2 is traveling on is divided into link 1613 and link 1614. In addition, the possible directions of travel for links 1611 and link 1614 are determined from the movements of other vehicles 1601 and 1602. The possible direction of travel 1622 for link 1613 can be inferred from the movement of vehicle 2. Then, as vehicle 2 moves to the position shown in Figure 16C, the recognition range of the parking space by vehicle 2 expands further, and partial lanes 1615, 1616, and 1617 are updated or newly added. Through these complementary processes, the lane structure of almost the entire parking lot shown in Figure 3 is generated.

[0112] In this way, vehicle 2 can estimate the overall road structure of the parking lot by driving only a part of it, without having to drive the entire parking lot. Understanding the road structure of the parking lot makes various things possible. For example, as shown in the driving plan unit 15 described later, it can suggest recommended routes for driving through the parking lot, such as searching for available parking spaces, or it can drive automatically. Furthermore, by managing the understood availability of parking spaces in relation to the road structure based on the recognition information from the external sensor group 4, it is possible to suggest routes that efficiently search for available parking spaces or to drive automatically.

[0113] Furthermore, by understanding the layout of the parking lot's pathways, it is possible to estimate the extent of the parking space rows. For example, in Figure 15B, only a portion of the parking spaces in the rows facing links 1521, 1524, and 1525 are recognized. By constructing the layout of the parking lot's pathways, it can be estimated that parking spaces exist in area 1541-1544. Knowing the extent of parking spaces in a row allows for determining whether a parking space is available by observing from a distance whether other vehicles are present, without having to physically go near the space to recognize it. Then, by associating the availability of parking spaces with the pathway structure, an efficient route for searching for available parking spaces can be suggested.

[0114] The generated parking lot map data set 35 may be stored in the storage unit 30 and read out from the storage unit 30 for use when driving through the same parking lot. Alternatively, the generated parking lot map data set 35 may be uploaded to a central system so that other vehicles can use it when driving through the same parking lot.

[0115] (Processing of the Driving Planning Unit 15) The processing performed by the Driving Planning Unit 15 will be explained with reference to the flowchart in Figure 17.

[0116] In step S1801, the driving plan unit 15 acquires the parking lot map data generated by the parking lot map generation unit 14.

[0117] Next, in step S1802, the driving plan unit 15 identifies the road area reachable from the vehicle 2's position based on information such as the link connection relationships and possible driving directions in the parking map data. For example, in Figure 16B, links 1613 and 1614 represent the road area reachable. In Figure 16C, these are the partial roads (links) 1615, 1617, 1611, 1613, and 1614.

[0118] In steps S1803 and S1804, the driving plan unit 15 checks the conditions for providing driving assistance in the parking lot. In this embodiment, since the parking lot map data is dynamically configured, the parking lot map is insufficient when the vehicle 2 enters the parking lot, and driving assistance in the parking lot cannot be provided. As the vehicle 2 drives through the parking lot, the parking lot map is dynamically configured, and at some point, it becomes possible to provide driving assistance. Therefore, it is desirable to repeatedly (for example, periodically) check at predetermined intervals whether the conditions necessary for providing driving assistance are met, and to notify the occupants that driving assistance is possible when the conditions for providing driving assistance are met. In step S1803, the distance that the vehicle 2 can continuously drive is calculated based on the reachable road area identified in step S1802, and if the calculated distance that can be driven is greater than or equal to a predetermined value, it is determined that the conditions for driving assistance are met. In step S1804, if a driving route that allows for circular driving back to a waypoint is detected within the parking lot, it is determined that the conditions for driving assistance are met. In the flowchart of Figure 17, if either step S1803 or step S1804 is met, the driving plan unit 15 generates a recommended driving route in step S1805. If neither condition is met, the process ends. For example, in the states of Figures 16A and 16B, the drivable distance is insufficient, and there is no driving route that allows for circular driving, so the conditions for driving assistance are not met. On the other hand, in the state of Figure 16C, there is a driving route that allows for circular driving (1615 → 1617 → 1611 → 1613), and sufficient driving distance can be secured, so it can be determined that the conditions for driving assistance are met.

[0119] After generating the recommended driving route, in step S1806, the driving planning unit 15 determines whether the vehicle 2 is in autonomous driving mode based on the information indicating the driving mode of the vehicle 2 included in the vehicle information data group 32. If the vehicle is in autonomous driving mode (Y in S1806), in step S1807, it generates driving control information to continue autonomous driving.

[0120] In step S1808, the driving plan unit 15 generates screen display information based on the information of the reachable road area calculated in step S1802 and the information of the recommended driving route calculated in step S1805. Then, in step S1809, the driving plan unit 15 stores the generated screen display information and driving control information in the storage unit 30 as a driving plan data group 36. The information output unit 16 outputs the screen display information to the HMI device group 8 and the driving control information to the actuator group 7.

[0121] Although not explicitly shown in Figure 17, the screen display information assists the occupant in driving within the parking lot in manual driving mode and also informs the occupant that it is possible to switch to automatic driving mode. The occupant understands the recommended driving route and the possibility of switching to automatic driving mode presented on the HMI device group 8, and issues instructions to the vehicle system 1 to switch to automatic driving mode as needed. The recommended driving route also presents the occupant with the route that vehicle 2 intends to travel in automatic driving mode. The electronic control unit 3 may also combine the recommended driving route with information on available parking spaces around vehicle 2 to generate a route that is highly likely to have available parking spaces.

[0122] Figure 18 shows an example of screen display information.

[0123] In the generated track structure, reachable track areas (links) are represented by solid lines, while unreachable or uncertain track areas are represented by dashed lines. The recommended driving route is represented by a thick solid line, and the route is set to follow links 1615, 1617, and 1611 in that order. The screen display information shown in Figure 18 includes a button 1901 for switching to autonomous driving mode. When autonomous driving assistance becomes available, button 1901 is activated, indicating to the occupants the availability of autonomous driving assistance, and the occupants can start autonomous driving assistance by operating button 1901.

[0124] As described above, by repeatedly checking the reachable road area of ​​dynamically generated parking map data and notifying the occupants that driving assistance is available when the conditions are sufficient to provide it, driving assistance for parking can be provided immediately after driving a portion of the parking lot, even in parking lots that have never been driven in before.

[0125] It should be noted that the present invention is not limited to the embodiments described above, but includes various modifications and equivalent configurations within the spirit of the attached claims. For example, the embodiments described above are described in detail for the purpose of clearly illustrating the present invention, and the present invention is not necessarily limited to having all the configurations described. Furthermore, some of the configurations of one embodiment may be replaced with those of another embodiment. Furthermore, configurations of other embodiments may be added to the configuration of one embodiment. Furthermore, some of the configurations of each embodiment may be added, deleted, or replaced with those of other embodiments.

[0126] For example, in the embodiment described above, it is assumed that each process performed by the electronic control unit 3 is executed in the same processing unit and storage unit, but it may also be executed in multiple different processing units and storage units. In that case, for example, processing software with a similar configuration may be installed in each of two or more storage units, and the processing may be divided and executed by each of the two or more processing units.

[0127] Furthermore, although each process performed by the electronic control unit 3 is realized by executing a predetermined program using a processor and RAM, it may be realized with proprietary hardware as needed. Also, in the above-described embodiment, the external sensor group 4, vehicle sensor group 5, actuator group 7, and HMI device group 8 are described as individual devices, but any two or more of them may be combined as needed to realize them.

[0128] Furthermore, each of the aforementioned configurations, functions, processing units, and processing means may be implemented in hardware, for example, by designing them as integrated circuits, or they may be implemented in software by having a processor interpret and execute programs that realize each function.

[0129] Information such as programs, tables, and files that implement each function can be stored in memory, hard disks, SSDs (Solid State Drives), or recording media such as IC cards, SD cards, and DVDs.

[0130] Furthermore, the drawings show control lines and information lines that are deemed necessary to explain the embodiments, and do not necessarily show all control lines and information lines included in actual products to which the present invention is applied. In practice, it can be assumed that almost all components are interconnected.

Claims

1. An electronic control device for generating a road structure in a parking lot, comprising: an information acquisition unit that acquires information about a plurality of other vehicles detected by a sensor mounted on the vehicle in the vicinity of the vehicle; a parking space row identification unit that identifies a plurality of partial parking space rows based on information about the position and orientation of other parked vehicles included in the other vehicle information; a partial road estimation unit that calculates a plurality of reference lines indicating the arrangement of parking space rows based on the arrangement of other parked vehicles in the partial parking space rows, and estimates a plurality of partial roads by shifting the calculated plurality of reference lines by a predetermined distance in the vertical direction; a parking lot map generation unit that generates a road structure around the vehicle by combining the estimated plurality of partial roads; and an information output unit that outputs the generated road structure.

2. An electronic control device according to claim 1, wherein the parking space row identification unit clusters the multiple other vehicles that are stopped based on the similarity of the distance and orientation between the multiple other vehicles that are stopped, identifies the direction in which the multiple other vehicles included in the cluster are lined up in the largest number of positions, and estimates the multiple partial parking space rows by re-clustering the multiple other vehicles included in the cluster based on the deviation of the positions of the other vehicles in the vertical direction of the identified direction.

3. An electronic control device according to claim 1, wherein the parking space row identification unit calculates the position and orientation of other vehicles that are stopped around the vehicle and its degree of confidence based on time-series data of information regarding the position and orientation of a plurality of other vehicles that are stopped, and estimates the plurality of partial parking space rows based on the position and orientation of other vehicles that are stopped and whose calculated degree of confidence is equal to or greater than a predetermined threshold.

4. An electronic control device according to claim 1, wherein the parking space row identification unit corrects the estimated parking space row based on the road structure generated by the parking map generation unit.

5. An electronic control device according to claim 1, wherein the parking lot map generation unit compares the trajectory of the position of another vehicle in motion, included in the other vehicle information, with the partial road to identify the partial road on which the other vehicle is traveling, identifies the possible directions of travel on the partial road based on the direction of movement of the other vehicle, and generates the road structure including the partial road on which the identified possible directions of travel are set.

6. An electronic control device according to claim 5, further comprising a driving plan unit that identifies a partial road reachable from the vehicle's position in a road structure around the vehicle, and generates information of the identified partial road as driving plan information, wherein the information output unit outputs the generated driving plan information.

7. An electronic control device according to claim 6, wherein the driving planning unit determines that driving assistance can be started when the portion of the road reachable from the position of the vehicle is equal to or greater than a predetermined distance threshold.

8. An electronic control device according to claim 6, wherein the driving planning unit determines that driving assistance can be started if a partial road reachable from the position of the vehicle is capable of circular driving.

9. A method for generating a parking map, performed by an electronic control device for generating a road structure in a parking lot, wherein the electronic control device comprises a computing device for performing predetermined calculation processing and a storage device connected to the computing device, and the method for generating a parking map is characterized by comprising: an information acquisition procedure for acquiring information about a plurality of other vehicles detected by a sensor mounted on a vehicle around the vehicle; a parking space row identification procedure for identifying a plurality of partial parking space rows based on information about the position and orientation of other vehicles that are stopped, which is included in the other vehicle information; a partial road estimation procedure for calculating a plurality of reference lines indicating the arrangement of parking space rows based on the arrangement of other vehicles stopped in the partial parking space rows, and estimating a plurality of partial roads by shifting the calculated plurality of reference lines by a predetermined distance in the vertical direction; a parking map generation procedure for generating a road structure around the vehicle by combining the estimated plurality of partial roads; and an information output procedure for outputting the generated road structure.

Images