Driving support device and computer program

Abstract

The present application provides a kind of driving support device and computer program, which can prevent the lane change not recommended when vehicle is driving, and can properly implement driving support. Specifically, using the map information containing lane shape, the lane moving mode candidates that vehicle can select when moving on the planned path are determined and obtained by lane network. Moreover, for the lane moving mode candidates obtained, the lane change start position and the lane change end position of the lane moving mode candidate with lane change are preferentially set at the position close to the destination, and the lane moving mode of the vehicle recommended when the vehicle moves is selected from the lane moving mode candidates considering the set lane change start position and lane change end position.

Description

Technical Field

[0001] This invention relates to a driving support device and a computer program for providing driving support for a vehicle. Background Technology

[0002] In order to provide appropriate driving support when the vehicle needs to change lanes from its current lane to another lane in the future, it is important to determine in advance when it is best to make a lane change.

[0003] Therefore, the following technology is proposed in International Publication No. 2017 / 159489: In generating an action plan for a vehicle to drive autonomously, the predetermined path of the vehicle is divided into multiple blocks, and a target lane is set for each block. Based on the set target lane, the action plan is generated by combining events executed in the vehicle, such as deceleration events, acceleration events, lane keeping events, and lane changing events, in the block.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: International Publication No. 2017 / 159489 (pages 10-14) Figure 6 ) Summary of the Invention

[0007] The problem the invention aims to solve

[0008] In the technology of Patent Document 1, the planned driving path of the vehicle is divided into multiple blocks, and lane change events are associated with these blocks. That is, although the timing of lane changes is determined on a block-by-block basis, to allow for flexibility in the plan, the specific timing of lane changes within a block is not predetermined (see paragraphs 0055-0058 of Patent Document 1). Figure 6 Furthermore, in the aforementioned Patent Document 1, the moment when a lane change actually occurs, i.e., when the vehicle is traveling in the block associated with the lane change event, is determined based on an assessment of the surrounding environment.

[0009] However, in action plans like those in Patent Document 1, where the timing of lane changes is predetermined in units of blocks with a certain width, it is possible for unrecommended lane changes to occur during actual vehicle movement. For example, Figure 28As shown, in generating an action plan for a predetermined driving path making a right turn at an intersection, suppose an action plan is generated where a lane change from the left lane to the center lane occurs in block A, which is closest to the intersection, and a lane change from the center lane to the right lane occurs in block B, which is closest to the intersection. In this case, if a vehicle traveling in block A changes lanes only when it is almost at the boundary of block A due to the presence of other vehicles, then in the shorter section of block B, lane changes in block A and block B may occur consecutively, or lane changes may occur near the intersection in block B. As a result, appropriate driving support may not be possible.

[0010] This invention was made to solve the aforementioned problems. Its purpose is to provide a driving support device and computer program that determines and obtains candidates for lane movement modes that a vehicle can select through a lane network. For each candidate, the lane change start position and lane change end position are preferentially set to be close to the destination. In this state, a recommended lane movement mode is selected from the candidates for lane movement modes. Therefore, it is possible to prevent unrecommended lane changes from occurring while the vehicle is in motion, thereby appropriately implementing driving support.

[0011] means for solving problems

[0012] To achieve the above objectives, the driving support device of the present invention includes: a driving predetermined path acquisition unit for acquiring a driving predetermined path for a vehicle; a lane network acquisition unit for acquiring a network representing lane movement that the vehicle can select, i.e., a lane network, based on map information including lane shapes, for the driving predetermined path; a candidate acquisition unit for determining and acquiring candidates for lane movement modes that the vehicle can select when moving on the driving predetermined path through the lane network; a lane change position setting unit for setting the lane movement mode that accompanies lane change among the candidates for lane movement modes acquired by the candidate acquisition unit, and the lane change start position for lane change end position for lane change end position, preferably at a position close to the destination; a recommended movement mode selection unit for selecting a recommended lane movement mode for the vehicle to move from the candidates for lane movement modes, considering the lane change start position and lane change end position set by the lane change position setting unit; and a driving support unit for providing driving support to enable the vehicle to move according to the lane movement mode selected by the recommended movement mode selection unit.

[0013] Furthermore, the computer program of the present invention is a program for generating support information for driving assistance implemented in a vehicle. Specifically, the computer program enables a computer to function as the following units: a driving predetermined path acquisition unit, which acquires a driving predetermined path for the vehicle; a lane network acquisition unit, which, based on map information including lane shapes, acquires a network representing lane movement that the vehicle can select for the driving predetermined path; a candidate acquisition unit, which determines and acquires candidates for lane movement modes that the vehicle can select when moving on the driving predetermined path through the lane network; a lane change position setting unit, which, for the candidates for lane movement modes acquired by the candidate acquisition unit that accompany lane changes, preferentially sets the lane change start position and the lane change end position to positions close to the destination; a recommended movement mode selection unit, which, considering the lane change start position and the lane change end position set by the lane change position setting unit, selects a recommended lane movement mode for the vehicle to move from the candidates for lane movement modes; and a driving assistance unit, which provides driving assistance for the vehicle to move according to the lane movement mode selected by the recommended movement mode selection unit.

[0014] Invention Effects

[0015] According to the driving support device and computer program of the present invention having the above-described structure, a candidate lane movement mode that the vehicle can select is determined and obtained through a lane network. For each candidate, the lane change start position (for initiating a lane change) and the lane change end position (for ending a lane change) are preferentially set to positions close to the destination. In this state, a recommended lane movement mode is selected from the candidate lane movement modes, thus enabling the appropriate selection of a lane movement mode for performing the recommended lane change. As a result, unrecommended lane changes can be prevented while the vehicle is driving, thereby enabling appropriate driving support. Attached Figure Description

[0016] Figure 1 This is a schematic structural diagram illustrating the driving support system of this embodiment.

[0017] Figure 2 This is a block diagram showing the structure of the driving support system in this embodiment.

[0018] Figure 3 This is a block diagram illustrating the navigation device of this embodiment.

[0019] Figure 4 This is a flowchart of the autonomous driving support procedure in this embodiment.

[0020] Figure 5It represents a map of the area where high-precision map information is obtained.

[0021] Figure 6 This is a flowchart of the subprocessing procedure for static driving trajectory generation.

[0022] Figure 7 This is a diagram representing an example of a vehicle's predetermined travel path.

[0023] Figure 8 It means to Figure 7 The diagram shows an example of a lane network constructed using a predetermined driving path.

[0024] Figure 9 This is a diagram illustrating an example of lane markings that show the correspondence between lanes contained in the road before the intersection and lanes contained in the road after the intersection.

[0025] Figure 10 It is a diagram representing the baseline path and candidate paths.

[0026] Figure 11 It is a diagram that shows the baseline path and alternative paths that take into account the set lane change locations.

[0027] Figure 12 It is a graph showing the relationship between the number of driving lanes and lane costs.

[0028] Figure 13 This is a diagram illustrating an example of how lane change costs are calculated.

[0029] Figure 14 This is a diagram illustrating an example of how lane change costs are calculated.

[0030] Figure 15 This diagram illustrates the method for calculating static driving trajectories within an intersection.

[0031] Figure 16 This is a flowchart of the sub-processing procedure for lane change location setting.

[0032] Figure 17 It is a diagram showing the range within which lanes can be changed.

[0033] Figure 18 This is a diagram that represents a temporary lane change position setting for the route.

[0034] Figure 19 It is a flowchart showing the process of modifying and finalizing the lane change position for the route setting.

[0035] Figure 20 It is a flowchart showing the process of modifying and finalizing the lane change position for the route setting.

[0036] Figure 21 It is a flowchart showing the process of modifying and finalizing the lane change position for the route setting.

[0037] Figure 22 This is a flowchart of the subprocessing procedure for generating and processing dynamic driving trajectories.

[0038] Figure 23 This is a diagram illustrating an example of an avoidance trajectory, which is one type of dynamic driving path.

[0039] Figure 24 This is a diagram illustrating an example of an avoidance trajectory, which is one type of dynamic driving path.

[0040] Figure 25 This is a diagram representing an example of a following trajectory, which is one type of dynamic driving trajectory.

[0041] Figure 26 This is a diagram representing an example of a following trajectory, which is one type of dynamic driving trajectory.

[0042] Figure 27 This is a flowchart of the sub-processing procedure for reflecting driving trajectory.

[0043] Figure 28 This is a diagram illustrating the problems with existing technologies. Detailed Implementation

[0044] Hereinafter, with reference to the accompanying drawings, a specific embodiment of the driving support device of the present invention as a navigation device 1 will be described in detail. First, using Figure 1 and Figure 2 This embodiment describes the general structure of the driving support system 2, which includes the navigation device 1. Figure 1 This is a schematic structural diagram showing the driving support system 2 of this embodiment. Figure 2 This is a block diagram showing the structure of the driving support system 2 in this embodiment.

[0045] like Figure 1 As shown, the driving support system 2 of this embodiment basically includes: a server device 4 provided by an information distribution center 3; and a navigation device 1 installed in the vehicle 5 to provide various support related to the autonomous driving of the vehicle 5. Furthermore, the server device 4 and the navigation device 1 are configured to exchange electronic data via a communication network 6. Alternatively, other in-vehicle devices installed in the vehicle 5 or vehicle control devices that perform controls related to the vehicle 5 may be used instead of the navigation device 1.

[0046] Among them, vehicle 5 is a vehicle that, in addition to manual driving based on the user's driving operation, can also drive with the support of automatic driving, which enables the vehicle to automatically drive along a pre-set path or road without the user's driving operation.

[0047] Furthermore, autonomous driving support can be implemented on all road sections, or only when the vehicle is traveling on a specific road section (e.g., a highway with exits / entries at the boundaries, regardless of whether there are people or not, or whether it is toll-free). In the following explanation, the autonomous driving zones for which vehicle autonomous driving support is implemented will include not only all road sections, including general roads and highways, but also parking lots, and autonomous driving support will be implemented essentially from the start to the end of the vehicle's journey. However, it is desirable that autonomous driving support is not necessarily implemented while the vehicle is traveling within an autonomous driving zone, but only when the user selects to implement autonomous driving support (e.g., turning the autonomous driving start button ON) and it is determined that driving based on autonomous driving support is possible. On the other hand, vehicle 5 can also be a vehicle that can only operate with autonomous driving support.

[0048] Furthermore, in vehicle control supported by autonomous driving, for example, the vehicle's current position, the lane it is traveling in, and the positions of surrounding obstacles are detected at any time. As described later, vehicle control of the steering wheel, drive source, brakes, etc., is automatically performed to travel along the driving trajectory generated by navigation device 1 and at a speed according to the same speed plan. In addition, in the autonomous driving-supported driving of this embodiment, lane changes and left and right turns can also be performed by driving through vehicle control based on the aforementioned autonomous driving support. However, special driving such as lane changes and left and right turns can also be performed manually without autonomous driving support.

[0049] On the other hand, the navigation device 1 is an in-vehicle unit mounted on the vehicle 5, which displays a map of the area surrounding the vehicle's location based on map data stored in the navigation device 1 or map data acquired from external sources, or displays the vehicle's current position on the map image, or provides movement guidance along a pre-set guidance path. In this embodiment, especially when the vehicle is driving with autonomous driving support, various support information related to autonomous driving support is generated. Such support information includes, for example, a recommended driving trajectory (including a recommended lane movement method) and a speed plan indicating the vehicle's speed. Further details regarding the navigation device 1 will be described later.

[0050] Furthermore, server device 4 performs route search based on requests from navigation device 1. Specifically, when a destination is set in navigation device 1 or when a route search is re-executed (route change), navigation device 1 sends information and a route search request, including the origin and destination, to server device 4 (however, in the case of a re-search, it is not necessarily necessary to send information related to the destination). Then, server device 4, upon receiving the route search request, uses its map information to perform a route search and determines a recommended route from the origin to the destination. It then sends the determined recommended route to navigation device 1, which is the request source. Navigation device 1 then provides information related to the received recommended route to the user, or sets the recommended route as a guide route and generates various support information related to autonomous driving support according to the guide route. Therefore, even if the map information held by navigation device 1 at the time of route search is an older version or navigation device 1 itself does not have map information, it is possible to provide an appropriate recommended route to the destination based on the latest version of map information held by server device 4.

[0051] Furthermore, in addition to the usual map information used for the aforementioned path search, server device 4 also possesses higher-precision map information, namely, high-precision map information. High-precision map information includes, for example, information related to road lane shapes (road shape or curvature in lane units, lane width, etc.) and lane lines drawn on the road (lane center line, lane boundary line, lane outer line, guide line, etc.). In addition to the above, it also includes information related to intersections, parking lots, etc. Moreover, server device 4 publishes high-precision map information based on requests from navigation device 1, and navigation device 1 uses the high-precision map information published from server device 4 to generate various support information related to autonomous driving support, as described later. Furthermore, while high-precision map information is generally map information that only includes roads (links) and their surroundings, it can also include map information that includes areas beyond the road's perimeter.

[0052] However, the path search processing described above does not necessarily have to be performed by server device 4. If the navigation device 1 has map information, it can also perform the processing. Furthermore, the high-precision map information can be pre-existing in the navigation device 1, rather than being published by server device 4.

[0053] Furthermore, the communication network 6 comprises multiple base stations deployed throughout the country and communication companies that control and manage each base station, interconnected via wired (fiber optic, ISDN, etc.) or wireless means. Each base station has a transceiver and antenna for communicating with the navigation device 1. The base station communicates wirelessly with the communication company and, as a terminal of the communication network 6, relays the communication of the navigation device 1 within the range (cell) reached by the radio waves from the base station to the server device 4.

[0054] Next, use Figure 2 The structure of the server device 4 in the driving support system 2 will be described in more detail. For example... Figure 2 As shown, the server device 4 includes: a server control unit 11, a server-side map DB12 connected to the server control unit 11 as an information recording unit, a high-precision map DB13, and a server-side communication device 14.

[0055] The server control unit 11 is a control unit (MCU, MPU, etc.) that performs overall control of the server device 4. It includes: a CPU 21 serving as both an arithmetic and control unit; and internal storage devices such as RAM 22 used as working memory during various arithmetic operations performed by the CPU 21, a ROM 23 storing control programs, and flash memory 24 storing programs read from the ROM 23. Furthermore, both the server control unit 11 and the ECU of the navigation device 1 (described later) have various units serving as processing algorithms.

[0056] On the other hand, the server-side map DB12 is a storage unit that stores server-side map information, which is the latest version of the map information registered based on input data or input operations from external sources. This server-side map information, primarily composed of road networks, consists of various information required for path searching, path guidance, and map display. For example, it includes network data representing nodes and road segments, road segment data related to roads (segments), node data related to nodes, intersection data related to intersections, location data related to facilities and other locations, map display data for displaying the map, search data for searching paths, and retrieval data for retrieving locations.

[0057] In addition, the high-precision map DB13 is a storage unit that stores map information with higher precision than the aforementioned server-side map information, namely high-precision map information 15. High-precision map information 15 specifically stores map information with more detailed information about roads, parking lots, etc., which are the objects to which vehicles travel. In this embodiment, it includes, for example, information related to the shape of road lanes (road shape, curvature, lane width, etc. in terms of lanes) and lane lines drawn on the road (lane center line, lane boundary line, lane outer line, guide line, etc.). Furthermore, the high-precision map DB13 records the following data: For each road segment constituting a road, shape interpolation point data is used to determine the width, slope, cant, embankment, and pavement condition of the road to which the segment belongs; shape interpolation point data is used to determine the shape of road segments between nodes (e.g., the shape of curves in curved roads); data representing merging sections, road structure, number of lanes, places where the number of lanes decreases, places where the width narrows, and intersections; for curves, data representing the radius of curvature, intersections, T-junctions, and the entrance and exit of curves; for road attributes, data representing downhill roads, uphill roads, etc.; for road types, in addition to data representing general roads such as national highways, county roads, and narrow streets, data representing toll roads such as national expressways, urban expressways, dedicated motor vehicle roads, general toll roads, and toll bridges. Especially in this embodiment, in addition to the number of road lanes, information is stored that determines the traffic category of each lane's direction of travel and the road connections (specifically, the correspondence between lanes contained in the road before and after an intersection). Furthermore, it also stores the speed limits set on the road. Additionally, high-precision map information is generally map information that only considers roads (segments) and their surroundings, but it can also include map information that includes areas beyond the road's perimeter. Furthermore, in Figure 2 In the example shown, although the server-side map information stored in the server-side map DB12 is different from the high-precision map information 15, the high-precision map information 15 can also be part of the server-side map information.

[0058] On the other hand, the server-side communication device 14 is a communication device used to communicate with the navigation devices 1 of each vehicle 5 via the communication network 6. In addition to the navigation devices 1, it can also receive traffic information composed of various information such as congestion information, restriction information, and traffic accident information sent by the Internet, traffic information centers, such as VICS (registered trademark: Vehicle Information and Communication System) centers.

[0059] Next, use Figure 3The general structure of the navigation device 1 mounted on vehicle 5 will be described. Figure 3 This is a block diagram illustrating the navigation device 1 of this embodiment.

[0060] like Figure 3 As shown, the navigation device 1 of this embodiment includes: a current position detection unit 31 that detects the current position of the vehicle equipped with the navigation device 1; a data recording unit 32 that records various data; a navigation ECU 33 that performs various calculations based on input information; an operation unit 34 that receives operations from the user; a liquid crystal display 35 that displays a map of the vehicle's surroundings and information related to the guidance path (pre-determined driving path of the vehicle) set by the navigation device 1; a speaker 36 that outputs sound guidance related to the path guidance; a DVD drive 37 that reads DVDs as storage media; and a communication module 38 that communicates with information centers such as detection centers and VICS centers. Furthermore, the navigation device 1 is connected to an external camera 39 and various sensors installed in the vehicle equipped with the navigation device 1 via an in-vehicle network such as CAN. Moreover, a vehicle control ECU 40 that performs various controls on the vehicle equipped with the navigation device 1 is connected in a manner capable of bidirectional communication.

[0061] The following sections will describe each component of the navigation device 1.

[0062] The current position detection unit 31 consists of a GPS 41, a vehicle speed sensor 42, a steering sensor 43, and a gyroscope sensor 44, and can detect the current vehicle position, orientation, vehicle speed, and current time. In particular, the vehicle speed sensor 42 is used to detect the vehicle's travel distance and speed. It generates pulses based on the rotation of the vehicle's drive wheels and outputs these pulse signals to the navigation ECU 33. The navigation ECU 33 then counts the generated pulses to calculate the rotational speed and travel distance of the drive wheels. Furthermore, the navigation device 1 does not necessarily need to have all four types of sensors; it can be configured to have only one or more of these sensors.

[0063] Additionally, the data recording unit 32 includes: a hard disk (not shown), which serves as an external storage device and recording medium; and a recording head (not shown), which acts as a drive for reading map information DB45, cache 46, specified programs, etc., recorded on the hard disk and for writing specified data to the hard disk. Alternatively, instead of a hard disk, the data recording unit 32 may also include a flash memory, memory card, CD, DVD, or other optical disc. Furthermore, in this embodiment, as described above, since the path to the destination is found in the server device 4, the map information DB45 can be omitted. Even when the map information DB45 is omitted, map information can still be obtained from the server device 4 as needed.

[0064] Among them, map information DB45 is, for example, a storage unit that stores road segment data related to roads (road segments), node data related to nodes, search data for processing route search or changes, facility data related to facilities, map display data for displaying the map, intersection data related to each intersection, and retrieval data for retrieving locations.

[0065] On the other hand, the cache 46 is a storage unit that stores the high-precision map information 15 previously published from the server device 4. The storage period can be appropriately set, for example, it can be a specified period after storage (e.g., 1 month), or it can be until the vehicle's ACC power supply (accessory power supply) is turned off. In addition, old data can be deleted sequentially after the amount of data stored in the cache 46 reaches its limit. Moreover, the navigation ECU 33 uses the high-precision map information 15 stored in the cache 46 to generate various support information related to autonomous driving support. Details will be described later.

[0066] On the other hand, the navigation ECU (Electronic Control Unit) 33 is an electronic control unit that performs overall control of the navigation device 1. This navigation ECU includes: a CPU 51 that serves as both a computing and control unit; and a RAM 52 that acts as working memory when the CPU 51 performs various calculations and stores path data, etc., during path searching; in addition to the control program, it also stores the autonomous driving support program (described later). Figure 4The navigation ECU 33 includes internal storage devices such as a ROM 53 and a flash memory 54 storing programs read from the ROM 53. Additionally, the navigation ECU 33 has various units that function as processing algorithms. For example, a driving predetermined path acquisition unit acquires a driving predetermined path for the vehicle. A lane network acquisition unit acquires a lane network for the driving predetermined path based on map information including lane shapes; this lane network represents the lane movement that the vehicle can choose. A candidate acquisition unit determines and acquires candidates for lane movement modes that the vehicle can choose when moving on the driving predetermined path using the lane network. A lane change position setting unit, for the candidates for lane movement modes acquired by the candidate acquisition unit, prioritizes setting the lane change start position (for lane changes) and the lane change end position (for lane changes) to positions close to the destination. A recommended movement mode selection unit considers the lane change start position and lane change end position set by the lane change position setting unit and selects a recommended lane movement mode for the vehicle from the candidates for lane movement modes. A driving support unit provides driving support to enable the vehicle to move according to the lane movement mode selected by the recommended movement mode selection unit.

[0067] The operation unit 34 is operated when the starting point of the journey and the destination point are input, and it has various operation switches (not shown) such as keys and buttons. Furthermore, the navigation ECU 33 controls the execution of various corresponding actions based on the switch signals output by pressing each switch. In addition, the operation unit 34 may also have a touch panel provided on the front surface of the LCD display 35. Additionally, it may include a microphone and a voice recognition device.

[0068] In addition, the LCD display 35 can display map images of roads, traffic information, operation guidance, operation menus, key guidance, guidance information along the guided route (driving the predetermined route), news, weather forecasts, timetables, emails, TV programs, etc. Alternatively, a HUD or HMD can be used instead of the LCD display 35.

[0069] In addition, the speaker 36 outputs audio guidance and traffic information guidance based on instructions from the navigation ECU 33 to guide driving along the guidance path (driving the predetermined path).

[0070] Additionally, the DVD drive 37 is a drive capable of reading data recorded on media such as DVDs or CDs. Furthermore, it can be used to play music or videos, update map information DB45, and perform other functions based on the read data. Alternatively, a card slot for reading and writing memory cards can be provided instead of the DVD drive 37.

[0071] Additionally, the communication module 38 is a communication device, such as a mobile phone or DCM, used to receive traffic information, detection information, weather information, etc., sent from traffic information centers, such as VICS centers or detection centers. It also includes a vehicle-to-vehicle communication device for communication between vehicles and a roadside communication device for communication with roadside equipment. Furthermore, it is used to send and receive route information or high-precision map information 15 retrieved by the server device 4 between the server device 4 and the server device 4.

[0072] Furthermore, the exterior camera 39, composed of a camera using a solid-state imaging element such as a CCD, is mounted above the front bumper of the vehicle, positioned at a predetermined angle downwards relative to the horizontal direction along its optical axis. When the vehicle is operating in the autonomous driving zone, the exterior camera 39 captures images of the area in front of the vehicle's direction of travel. The navigation ECU 33 processes the captured images to detect lane lines drawn on the road surface, other vehicles, and other obstacles, and generates various support information related to autonomous driving based on the detection results. For example, when an obstacle is detected, a new driving trajectory is generated to avoid or follow the obstacle. Alternatively, the exterior camera 39 can be configured to be positioned behind or to the side of the vehicle, other than in front of it. Furthermore, sensors such as millimeter-wave radar and laser sensors, vehicle-to-vehicle communication, and road-to-road communication can be used instead of cameras as the obstacle detection unit.

[0073] Furthermore, the vehicle control ECU 40 is an electronic control unit that controls the vehicle equipped with the navigation device 1. The vehicle control ECU 40 is connected to various drive components of the vehicle, such as the steering wheel, brakes, and accelerator. In this embodiment, especially after the vehicle begins autonomous driving support, autonomous driving support is implemented by controlling these various drive components. Additionally, if the user overtakes the vehicle during autonomous driving support, this overtaking situation will be detected.

[0074] After driving begins, the navigation ECU 33 sends various support information related to the autonomous driving support generated by the navigation device 1 to the vehicle control ECU 40 via CAN. The vehicle control ECU 40 then uses the received support information to implement autonomous driving support after driving begins. This support information includes, for example, a recommended driving trajectory and a speed plan indicating the vehicle's speed.

[0075] Next, based on Figure 4 The autonomous driving support program executed by the CPU 51 in the navigation device 1 of this embodiment having the above-described structure will be described. Figure 4This is a flowchart of the autonomous driving support procedure of this embodiment. The autonomous driving support procedure is executed after the vehicle's ACC (accessory power supply) is turned on and the vehicle begins autonomous driving support operation, and implements the autonomous driving support-based driving procedure according to the support information generated by the navigation device 1. Furthermore, in the following... Figure 4 , Figure 6 , Figure 16 , Figure 22 as well as Figure 27 The program shown in the flowchart is stored in the RAM52 and ROM53 of the navigation device 1 and executed by the CPU51.

[0076] First, in step S1 of the autonomous driving support program, CPU 51 obtains the predetermined path (hereinafter referred to as the driving predetermined path) for the vehicle's future travel. Furthermore, in the vehicle's driving predetermined path, when a guidance path is set in the navigation device 1, the path from the vehicle's current position to its destination within the currently set guidance path in the navigation device 1 is used as the driving predetermined path. On the other hand, when no guidance path is set in the navigation device 1, the path traveled from the vehicle's current position along the road can also be used as the driving predetermined path.

[0077] Furthermore, the guided path is a recommended path from the origin to the destination set by the navigation device 1, which in this embodiment is specifically searched by the server device 4. When searching for a recommended path, the CPU 51 first sends a path search request to the server device 4. The path search request includes information such as the terminal ID of the navigation device 1 that sent the path search request, the origin (e.g., the current location of the vehicle), and the destination. It should be noted that the destination information is not necessarily required during a re-search. Then, the CPU 51 receives search path information sent from the server device 4 according to the path search request. The search path information is information (e.g., a list of road segments included in the recommended path) used by the server device 4 to determine the recommended path (central path) from the origin to the destination searched by the server device 4 based on the sent path search request and using the latest version of map information. For example, the search is performed using the well-known Dijkstra method. The CPU 51 then sets the received recommended path as the guided path for the navigation device 1.

[0078] Next, in S2, CPU 51 acquires high-precision map information 15, targeting an area within a predetermined distance along the predetermined driving path obtained in S1, starting from the vehicle's current position. For example, the predetermined driving path contained in the secondary grid where the vehicle is currently located can be used as the target for acquiring high-precision map information 15. However, the area that becomes the target for acquiring high-precision map information 15 can be appropriately changed; for example, high-precision map information 15 can also be acquired for an area within 3km along the predetermined driving path starting from the vehicle's current position. Alternatively, the entire predetermined driving path can be used as the target for acquiring high-precision map information 15.

[0079] Among them, such as Figure 5 As shown, the high-precision map information 15 is divided into rectangular shapes (e.g., 500m × 1km) and stored in the high-precision map DB13 of the server device 4. Therefore, for example... Figure 5 As shown, when a predetermined driving path 61 is obtained, high-precision map information 15 is acquired for regions 62-64. Regions 62-64 contain portions of the predetermined driving path 61 located within a secondary grid, including the vehicle's current position. The high-precision map information 15 includes information such as lane shape, lane width, and lane lines drawn on the road (lane center line, lane boundary line, lane outer line, guide line, etc.). Additionally, it includes information related to intersections and parking lots.

[0080] Furthermore, although the high-precision map information 15 is primarily obtained from the server device 4, if high-precision map information 15 for a region is already stored in the cache 46, it is obtained from the cache 46. Additionally, the high-precision map information 15 obtained from the server device 4 is temporarily stored in the cache 46.

[0081] Then, in S3, CPU51 performs the static driving trajectory generation process described later. Figure 6The static driving trajectory generation process is based on the vehicle's predetermined driving path and the high-precision map information 15 obtained in S2 above. It generates a recommended driving trajectory, i.e., a static driving trajectory, for the roads included in the predetermined driving path. Specifically, the CPU 51 determines the recommended driving trajectory as the static driving trajectory on a per-lane basis, using the lanes included in the predetermined driving path as the unit. Furthermore, as described later, the static driving trajectory is generated using the interval from the vehicle's current position along the direction of travel up to a predetermined distance (e.g., within the secondary grid where the vehicle is currently located or the entire interval up to the destination). Additionally, although the predetermined distance can be appropriately changed, the static driving trajectory is generated using an area that at least includes the region outside the detection range (the area where the external camera 39 or other sensors can detect the road conditions around the vehicle).

[0082] Next, in S4, CPU51 generates a speed plan for the vehicle traveling on the static driving trajectory generated in S3, based on the high-precision map information 15 obtained in S2 above. For example, it calculates the recommended driving speed for the vehicle when traveling on the static driving trajectory, taking into account speed limit information and speed change locations on the predetermined driving path (such as intersections, curves, road crossings, pedestrian crossings, etc.).

[0083] Then, the speed plan generated in S4 above is stored as support information for autonomous driving support in flash memory 54, etc. Alternatively, an acceleration plan representing the acceleration / deceleration of the vehicle required to realize the speed plan generated in S4 above can also be generated as support information for autonomous driving support.

[0084] Next, in S5, CPU51 performs image processing on the images captured by the external camera 39, specifically determining whether there are factors affecting the vehicle's movement in the vicinity of the vehicle as the surrounding road conditions. The "factors affecting the vehicle's movement" that are the subjects of judgment in S5 are defined as dynamic factors that change in real time, excluding static factors based on road structure. Examples include other vehicles traveling or parked ahead of the vehicle, pedestrians ahead of the vehicle's direction of travel, and construction zones ahead of the vehicle's direction of travel. On the other hand, intersections, curves, crossings, merging sections, and lane reduction sections are excluded. Furthermore, even if other vehicles, pedestrians, or construction zones are present, those that are unlikely to overlap with the vehicle's future trajectory (e.g., located far from the vehicle's future trajectory) are excluded from the "factors affecting the vehicle's movement." Additionally, as the unit for detecting factors that may affect the vehicle's movement, sensors such as millimeter-wave radar and laser sensors, vehicle-to-vehicle communication, and road-to-road communication can be used instead of cameras.

[0085] Furthermore, if it is determined that there are factors in the vicinity of the vehicle that affect its operation ("Yes" in S5), the process proceeds to S6. Conversely, if it is determined that there are no factors in the vicinity of the vehicle that affect its operation ("No" in S5), the process proceeds to S9.

[0086] In S6, CPU51 performs the dynamic driving trajectory generation process described later. Figure 22 The dynamic driving trajectory generation process generates a new trajectory as the dynamic driving trajectory, which is used to avoid or follow the "factors affecting the driving of this vehicle" detected in S5 above from the vehicle's current position and return to the static driving trajectory. Furthermore, as described later, the dynamic driving trajectory is generated based on the interval containing the "factors affecting the driving of this vehicle". The length of the interval varies depending on the content of the factor. For example, if the "factor affecting the driving of this vehicle" is another vehicle (the vehicle in front) driving in front of the vehicle, as an example, a trajectory is generated that involves changing lanes to the right to overtake the vehicle in front, then changing lanes to the left to return to the original lane. Furthermore, since the dynamic driving trajectory is generated based on the road conditions around the vehicle obtained by the external camera 39 or other sensors, the area where the dynamic driving trajectory is generated is at least within the range (detection range) where the road conditions around the vehicle can be detected by the external camera 39 or other sensors.

[0087] Next, in S7, CPU51 performs the driving trajectory reflection processing described later. Figure 27 The driving trajectory reflection process involves reflecting the newly generated dynamic driving trajectory in S6 onto the static driving trajectory generated in S3. Specifically, for the section from the vehicle's current position to the end of the interval containing "factors affecting the vehicle's driving," the cost of the static driving trajectory and at least one dynamic driving trajectory is calculated, and the driving trajectory with the lowest cost is selected. As a result, a portion of the static driving trajectory is replaced with the dynamic driving trajectory as needed. Furthermore, depending on the situation, sometimes the dynamic driving trajectory replacement is not performed; that is, even if the dynamic driving trajectory is reflected, the static driving trajectory generated in S3 may not change. Moreover, if the dynamic and static driving trajectories are the same, even if replacement is performed, the static driving trajectory generated in S3 may not change.

[0088] Next, in S8, CPU51 corrects the vehicle speed plan generated in S4 based on the content of the reflected dynamic driving trajectory, which was reflected in S7, for the static driving trajectory. Alternatively, if reflecting the dynamic driving trajectory does not change the static driving trajectory generated in S3, the processing in S8 can be omitted.

[0089] Next, in S9, CPU 51 calculates the control quantities used to make the vehicle travel at a speed according to the speed plan generated in S4 (or the revised plan if the speed plan was revised in S8) on the static driving trajectory generated in S3 (or the reflected trajectory if the dynamic driving trajectory was reflected in S7). Specifically, the control quantities for the accelerator, brakes, gear, and steering wheel are calculated respectively. Furthermore, the processing in S9 and S10 can also be performed by the vehicle control ECU 40, which controls the vehicle, instead of by the navigation device 1.

[0090] Then, in S10, the CPU 51 reflects the control quantity calculated in S9. Specifically, the calculated control quantity is sent to the vehicle control ECU 40 via CAN. The vehicle control ECU 40 performs various vehicle controls, including accelerator, brake, gear shift, and steering wheel, based on the received control quantity. As a result, driving support control is possible, allowing the vehicle to travel at the speed generated in S3 according to the speed plan generated in S4 (or the revised plan if the speed plan was revised in S8) and on the static driving trajectory generated in S3 (or the reflected trajectory if the dynamic driving trajectory was reflected in S7).

[0091] Next, in S11, CPU51 determines whether the vehicle has traveled a certain distance after generating the static driving trajectory in S3. For example, the certain distance is 1 km.

[0092] Furthermore, if it is determined in S3 that the vehicle has traveled a certain distance after generating the static driving trajectory (in S11, this is "yes"), the process returns to S1. Then, the static driving trajectory is generated again (S1 to S4) using the interval within a predetermined distance along the predetermined driving path starting from the vehicle's current position as the target. In addition, in this embodiment, whenever the vehicle travels a certain distance (e.g., 1 km), the static driving trajectory is repeatedly generated using the interval within a predetermined distance along the predetermined driving path starting from the vehicle's current position as the target. However, if the distance to the destination is short, the static driving trajectory to the destination can also be generated all at once at the start of the journey.

[0093] On the other hand, if it is determined in S3 that the vehicle has not traveled a certain distance after generating the static driving trajectory (in S11, it is "No"), then it is determined whether to terminate the support driving based on autonomous driving support (S12). In addition to reaching the destination, the situation where the support driving based on autonomous driving support is terminated may also be when the user operates the control panel set on the vehicle, or intentionally releases (overtakes control) the support driving based on autonomous driving support by operating the steering wheel, braking, etc.

[0094] Then, if it is determined that the support driving based on autonomous driving support has ended ("Yes" in S12), the autonomous driving support procedure is terminated. Conversely, if it is determined that the support driving based on autonomous driving support should continue ("No" in S12), the process returns to S5.

[0095] Next, based on Figure 6 The subprocesses of the static driving trajectory generation process executed in S3 above will be explained. Figure 6 This is a flowchart of the subprocessing procedure for static driving trajectory generation.

[0096] First, in S21, the CPU 51 acquires the vehicle's current position detected by the current position detection unit 31. Furthermore, the vehicle's current position is preferably determined in detail using, for example, high-precision GPS information or high-precision positioning technology. High-precision positioning technology refers to a technique that detects white lines and road surface markings captured by a camera installed on the vehicle using image recognition, and then matches the detected white lines and road surface markings with, for example, high-precision map information 15, thereby enabling the detection of the driving lane and the vehicle's precise position. Moreover, when the vehicle is traveling on a road consisting of multiple lanes, the lane in which the vehicle is traveling is also determined.

[0097] Next, in S22, CPU51, based on the high-precision map information 15 obtained in S2 above, acquires lane shape, lane line information, and information related to intersections, targeting the interval of the static driving trajectory ahead of the vehicle's direction of travel (e.g., within a secondary grid including the vehicle's current position). Furthermore, the lane shape and lane line information acquired in S22 includes the number of lanes, lane width, the location and method of lane increases or decreases if there are changes in the number of lanes, and information on the connection of passage segments or roads in the direction of travel for each lane (specifically, the correspondence between lanes included in the road before the intersection and lanes included in the road after the intersection). Additionally, as information related to intersections, besides the shape of the intersection, it also includes information about the position and shape of objects placed on the ground at the intersection. Moreover, among the "objects placed on the ground at the intersection," in addition to road surface displays such as guide lines (guide white lines) and diamond-shaped guide strips (diamond-shaped markers) placed in the center of the intersection, there are structures such as poles.

[0098] Next, in S23, CPU51 constructs a lane network based on the lane shape and lane line information obtained in S22 above, using the interval of the static driving trajectory ahead of the vehicle's direction of travel as the object. Here, the lane network represents the network of lanes that the vehicle can choose to move within.

[0099] As an example of constructing a lane network in S23 above, for example, using vehicles in Figure 7 The following example illustrates the situation of traveling on a predetermined route. Figure 7 The planned driving route shown is a path from the vehicle's current position: go straight, turn right at the next intersection 71, turn right again at the next intersection 72, and turn left at the next intersection 73. Figure 7 In the planned driving path shown, for example, when turning right at intersection 71, one can enter either the right-hand lane or the left-hand lane. However, since a right turn is required at the next intersection 72, one must move to the rightmost lane when entering intersection 72. Similarly, when turning right at intersection 72, one can enter either the right-hand lane or the left-hand lane. However, since a left turn is required at the next intersection 73, one must move to the leftmost lane when entering intersection 73. Figure 8 The diagram shows a lane network constructed using such lane-moving sections as objects.

[0100] like Figure 8As shown, the lane network divides the area of ​​the static driving trajectory ahead of the vehicle's direction of travel into multiple segments (groups). Specifically, the division is based on the entry position of the intersection, the exit position of the intersection, and the position of lane entry / exit. Then, nodes (hereinafter referred to as lane nodes) 75 are set for each lane located at the boundary of each segment. Furthermore, road segments (hereinafter referred to as road segments) 76 connecting the lane nodes 75 are set.

[0101] Furthermore, the aforementioned lane network includes the following information: determining the correspondence between lanes included in the road before the intersection and lanes included in the road after the intersection, based particularly on the connection between lane nodes and road segments at intersections; that is, information on lanes that can be moved after the intersection compared to lanes before the intersection. Specifically, it shows the lanes that a vehicle can move between lane nodes set on the road before the intersection and lane nodes connected by road segments set on the road after the intersection.

[0102] To generate such a lane network, in the high-precision map information 15, for each road connecting to the intersection, lane markers representing the correspondence between lanes are set and stored for each combination of roads entering and leaving the intersection. For example, in Figure 9 The diagram shows lane markings for entering an intersection from the right-hand road and leaving onto the upper road, lane markings for entering an intersection from the right-hand road and leaving onto the left road, and lane markings for entering an intersection from the right-hand road and leaving onto the lower road. Furthermore, it shows lanes whose lane markings are set to "1" in the road before the intersection and lanes whose lane markings are set to "1" in the road after the intersection, i.e., lanes that can move before and after the intersection. When constructing the lane network in S23, CPU51 refers to the lane markings to form connections between lane nodes and road segments in the intersection.

[0103] Next, in S24, CPU51 sets the starting lane for the vehicle to move from the lane node located at the starting point of the lane network, and sets the target lane for the vehicle to move from the lane node located at the ending point of the lane network, based on the lane network constructed in S23 above. Furthermore, if the starting point of the lane network is a one-way multi-lane road, the lane node corresponding to the lane the vehicle is currently in becomes the starting lane. On the other hand, if the ending point of the lane network is a one-way multi-lane road, the lane node corresponding to the leftmost lane (in the case of left-hand traffic) becomes the target lane.

[0104] Then, in S25, CPU51, referring to the lane network constructed in S23 above, searches for a continuous path from the starting lane to the target lane and derives a path (hereinafter referred to as the baseline path). For example, Dijkstra's method can be used to search for the path from the target lane side. However, search methods other than Dijkstra's method can also be used as long as a continuous path from the starting lane to the target lane can be found. In addition, the baseline path derived in S25 above does not necessarily have to be the optimal path; any path that is continuous from the starting lane to the target lane is acceptable.

[0105] Next, in S26, CPU51, based on the reference path determined in S25 above, derives other paths (hereinafter referred to as candidate paths) that continuously connect the starting lane to the target lane. For example, in the case of reaching a section where lane changing is possible or an intersection where left or right turns are possible from the starting lane along the reference path, a new path different from the reference path is generated by branching. The result is as follows: Figure 10 As shown, a baseline path and one or more candidate paths are generated.

[0106] Then, in S27, CPU51 executes the lane change position setting process described later. Figure 16 The lane change position setting process involves setting the specific lane change position (hereinafter referred to as the lane change position) for the base path and candidate path generated in S25 and S26 that are accompanied by lane changes. Furthermore, if the base path and candidate path generated in S25 and S26 are paths that do not involve any lane changes, the processing in S27 can be omitted.

[0107] Next, in S28, CPU 51 adds lane nodes 75 to the lane change locations set in S27, based on the baseline and candidate paths generated in S25 and S26. These lane change locations include the lane change start point and the lane change end point. In S28, lane nodes 75 are added to both the lane change start point and the lane change end point. Furthermore, along with the addition of lane nodes, road segments 76 connecting the lane nodes 75 are also added. Additionally, road segments outside the lane change locations are essentially set to straight lines (along the shape of the lane). The result is as follows: Figure 11 As shown, the baseline path and candidate path illustrate a more detailed lane movement pattern after the set lane change position is determined in S27 above. Furthermore, if the baseline path and candidate path generated in S25 and S26 above are paths that do not involve any lane changes, the processing in S28 can be omitted.

[0108] Next, in S29, CPU51 considers the set lane change positions for the baseline and candidate paths generated in S25 and S26, and for which lane change positions were set in S27 and lane nodes and road segments were added in S28, and calculates the total lane cost for each path. Then, it compares the total lane cost of each path and determines the path with the smallest total lane cost as the recommended lane movement method for the vehicle.

[0109] In this process, a lane cost is assigned to each road segment 76, and in S29, the total lane cost of all road segments 76 included in each path is calculated and compared. Furthermore, the lane cost assigned to each road segment 76 is based on the length of each road segment 76. Then, the base value is corrected under the following condition (1). Moreover, for paths involving lane changes, the cost calculated under condition (2) (hereinafter referred to as lane change cost) is added to the total lane cost, depending on the location or number of lane changes. Additionally, road segments 76 within the same section (group) are essentially considered to have the same length (i.e., the increase in distance caused by lane changes is ignored). Furthermore, the length of road segments 76 within intersections is considered to be 0 or a fixed value.

[0110] (1) For lane costs of road segments consisting of multiple lanes, a coefficient is multiplied from the baseline value based on the position of the driving lane. Specifically, lane segments traveling in the overtaking lane are adjusted so that their lane costs are higher than those traveling in the driving lane. As a result, paths with longer distances in the overtaking lane will have higher total lane costs, making them less likely to be selected as recommended lane movement modes. Paths with shorter distances in the overtaking lane are preferred as recommended lane movement modes. In addition, for roads that do not distinguish between driving lanes and overtaking lanes, in countries where driving is on the left, the adjustment is made so that lane costs are higher for lane segments traveling in the rightmost lane. For example, ... Figure 12 As shown, in a one-way three-lane road, the lane cost is multiplied by 1.0 for road segments traveling in the leftmost lane 81, by 1.1 for road segments traveling in the middle lane 82, and by 1.2 for road segments traveling in the rightmost lane 83. Conversely, in countries where traffic travels on the right, the lane cost is adjusted in such a way that the lane cost is higher the further to the left of the road segment.

[0111] (2) When calculating the lane change cost, the recommended position for performing the lane change is first obtained as the lane change recommendation position. In this embodiment, the lane change included in the path is based on the premise that the lane change is performed to pass through the fork point, and the lane change recommendation position is set based on the fork point that is the factor for the lane change. Specifically, the lane change recommendation position is determined by the road type of the road entering the fork point (i.e., the road type of the road for which the lane change is performed). For example, it is set at a position 2000m before the fork point on an intercity expressway. It is set at a position 1000m before the fork point on an urban expressway. It is set at a position 700m before the fork point on other general roads or narrow streets. Moreover, with the distance from the fork point to the lane change recommendation position set as L, and the distance from the fork point to the lane change position set in S27 above (more specifically, the lane change start point) set as X, the lane change cost C is calculated by the following formula (a).

[0112] C = |1 - X / L| × K (constant) ····(a)

[0113] Additionally, the value of K can be set appropriately, for example, to 200. For paths containing multiple lane changes, the lane change cost described above is calculated for each lane change and then added to the total lane cost.

[0114] For example, Figure 13 As shown, taking a path that involves two lane changes before becoming a right-turn target as an example, the method for calculating lane change costs is explained. Figure 13 In the example shown, for the first lane change far from the fork in the road, the cost is "0" because the distance from the fork to the lane change location is 700m, and the distance from the fork to the recommended lane change location is also 700m. On the other hand, for the second lane change closer to the fork, the cost is approximately "86" because the distance from the fork to the lane change location is 400m, and the distance from the fork to the recommended lane change location is 700m. Therefore, for... Figure 13 The path shown includes "0" and "86" as lane change costs added to the total lane cost. According to equation (a) above, on the path, the closer the lane change position set in S27 is to the recommended lane change position, the lower the lane change cost, and therefore it is preferentially selected as the recommended lane movement method. Furthermore, since paths with fewer lane changes result in fewer added lane costs, they are also preferentially selected as the recommended lane movement method.

[0115] Then, with Figure 13Taking the path shown as an example, the calculation method for the total lane cost is explained. Regarding the range from the branch point to 1000m before the branch point, the cost from 1000m to 700m before the branch point is 300m × 1.0 = 300; the cost from 700m to 400m before the branch point is 300m × 1.1 = 330; and the cost from 400m before the branch point to the branch point is 400m × 1.2 = 480. Then, if the lane change cost of "0" and "86" mentioned above is added to these lane costs, the final total lane cost is "1196". Furthermore, the total lane cost is actually calculated not only for the range from the branch point to 1000m before the branch point, but also for the entirety of the baseline path and candidate paths generated in S25 and S26 above.

[0116] On the other hand, with Figure 13 Other paths within the same interval Figure 14 Taking the path shown as an example, the calculation method for the total lane cost is explained. Regarding the range from the fork in the road to 1000m before the fork, the cost from 1000m to 400m before the fork is 600m × 1.0 = 600; the cost from 400m to 200m before the fork is 200m × 1.1 = 220; and the cost from 200m before the fork to the fork is 200m × 1.2 = 240. Additionally, in... Figure 14 In the example shown, for the first lane change far from the fork in the road, the distance from the fork to the lane change location is 400m, and the distance from the fork to the recommended lane change location is 700m, so the lane change cost is "86". On the other hand, for the second lane change closer to the fork in the road, the distance from the fork to the lane change location is 200m, and the distance from the fork to the recommended lane change location is 700m, so the lane change cost is approximately "143". Moreover, if the lane change costs of "86" and "143" are added to the total lane cost, the final total lane cost is "1269".

[0117] Therefore, Figure 13 The path shown and Figure 14 The paths shown are compared for the range from the bifurcation point to 1000m before the bifurcation point. Figure 13 The total lane cost of the shown path is relatively low, therefore it means... Figure 13 The path shown is more likely to be selected as the recommended lane movement method.

[0118] Next, in S30, CPU51 calculates a recommended driving trajectory based on the path selected in S29 above, especially for the sections (groups) where lane change positions were set in S27 above, when the vehicle moves into a lane. Alternatively, if the path selected in S29 is a path that does not involve any lane changes, the processing in S30 can be omitted.

[0119] Specifically, CPU51 uses map information such as the lane change location set in S27 above to calculate the driving trajectory. For example, based on the vehicle's speed (set as the speed limit for the road) and lane width, it calculates the lateral acceleration (lateral G) generated when the vehicle changes lanes. Using a value where the lateral G does not exceed the upper limit (e.g., 0.2G), meaning it will not hinder autonomous driving support or cause discomfort to the vehicle's occupants, it uses a clothoid curve to calculate the trajectory connecting the start and end points of the lane change as smoothly as possible. Furthermore, a clothoid curve is a curve depicted by the vehicle's trajectory when the steering wheel is turned at a certain speed and angular velocity.

[0120] Next, in S31, CPU51 calculates a recommended driving trajectory, especially for sections (groups) within intersections, based on the path selected in S29 above, when the vehicle moves within a lane. Alternatively, if the path selected in S29 does not pass through any intersections, the processing in S31 can be omitted.

[0121] For example, in Figure 15 This example illustrates the calculation of a driving trajectory for a segment (group) within an intersection where the vehicle's path is defined as entering from the rightmost lane and exiting from the leftmost lane. First, the CPU51 marks the positions the vehicle should pass through within the intersection. Specifically, it marks the entry position within the entry lane, the guide line within the intersection (only when guide lines are present), the exit position from the intersection, and the exit lane from the intersection. If there is a diamond-shaped guide strip (diamond mark) in the center of the intersection, the diamond mark is also considered for marking. Furthermore, the curve passing through all marked marks is calculated as the driving trajectory. More specifically, after connecting the marks with spline curves, a spiral curve approximating the connected curve is calculated as the driving trajectory. It should be noted that the spiral curve is the curve depicted by the vehicle's trajectory when the vehicle rotates the steering wheel at a certain speed and angular velocity.

[0122] Then, in S32, CPU51 generates a static driving trajectory that recommends vehicle travel for the roads included in the predetermined driving path by connecting the driving trajectories calculated in S30 and S31. Furthermore, for sections that are neither lane-changing sections nor sections within intersections, the trajectory passing through the center of the lane is used as the recommended vehicle travel trajectory.

[0123] Then, the static driving trajectory generated in S32 above is stored in flash memory 54, etc., as support information for autonomous driving support.

[0124] Next, based on Figure 16 The subprocesses of the lane change position setting process performed in S27 above will be explained. Figure 16 This is a flowchart of the sub-processing procedure for lane change location setting.

[0125] It should be noted that the following lane change position setting process applies to each base path and candidate path generated in S25 and S26 above, and is implemented for each "fork point (fork point accompanying lane change)" contained in each path. Furthermore, as mentioned above, in this embodiment, the lane change included in the path is assumed to be a lane change performed to pass through a fork point. Therefore, for a path with multiple "fork points where lane changes are performed to pass through," the process is performed separately for each of the multiple fork points. Moreover, for fork points where multiple lane changes are performed to pass through, the process is further implemented each time multiple lane changes are performed.

[0126] First, in S41, CPU51 sets the range within which lane changes can be performed on the bifurcation point that is being processed, i.e., the lane changeable range. Specifically, referring to the lane network constructed in S23 above, the lane node 75 that contains the end point of the segment (group) with the corresponding lane change is set as the end point of the lane changeable range. On the other hand, the start point of the lane changeable range is the lane node closest to the originating side where the lane change can be performed.

[0127] For example, Figure 17 As shown in the left figure, for a path that involves a two-lane lane change from the leftmost lane to the right in the section (group) closest to the fork point where a right turn is possible, lane node X1, which includes the end of the section (group) where the lane change occurs, becomes the end of the lane change range. On the other hand, lane node X2, which is closest to the starting point and allows a two-lane lane change from the leftmost lane to the right, becomes the starting point of the lane change range.

[0128] On the other hand, such as Figure 17As shown in the middle diagram, for a path where a lane change occurs from the leftmost lane to the right in the section (group) second closest to the fork point where a right turn is possible, and a lane change occurs from the center lane to the right in the section (group) closest to the fork point, firstly, lane node X3, which includes the end point of the section (group) where the first lane change occurs, becomes the end point of the first lane change range. On the other hand, lane node X4, which allows a lane change from the leftmost lane to the right and is closest to the starting point, becomes the starting point of the first lane change range. However, the starting point of the first lane change range can also be set as the lane node that includes the starting point of the section (group) where the lane change occurs. Furthermore, lane node X5, which includes the end point of the section (group) where the second lane change occurs, becomes the end point of the second lane change range. On the other hand, lane node X3, which allows a lane change from the center lane to the right and is closest to the starting point, becomes the starting point of the second lane change range.

[0129] In addition, such as Figure 17 As shown in the right figure, for a path where a lane change occurs from the leftmost lane to the right in the section (group) closest to the branch point where a right turn is desired, and a lane change occurs from the center lane to the right in the section (group) closest to the branch point, firstly, lane node X6, which includes the end point of the section (group) where the first lane change occurs, becomes the end point of the first lane change range. On the other hand, lane node X7, which allows a lane change from the leftmost lane to the right and is closest to the starting point, becomes the starting point of the first lane change range. Furthermore, lane node X8, which includes the end point of the section (group) where the second lane change occurs, becomes the end point of the second lane change range. On the other hand, lane node X9, which allows a lane change from the center lane to the right and is closest to the starting point, becomes the starting point of the second lane change range. The determination of lane nodes X1 to X9 is based on the lane network constructed in S23 above.

[0130] Next, in S42, CPU51 uses map information and vehicle information around the bifurcation point of the processing object to calculate the distance required for lane changing (hereinafter referred to as the lane change distance). The lane change distance is the distance along the direction of travel (the distance along the length of the road). For example, based on the vehicle's speed (set as the speed limit for the road) and lane width, the lateral acceleration (lateral G) generated when the vehicle changes lanes is calculated. Using a spiral curve as a condition, where the lateral G does not exceed an upper limit (e.g., 0.2G) that will not impede autonomous driving support or cause discomfort to the vehicle's occupants, a trajectory that is as smooth as possible and minimizes the distance required for lane changing is calculated. Furthermore, a spiral curve is a curve depicted by the trajectory of a vehicle when the steering wheel is turned at a certain speed and angular velocity. The lane change distance is then calculated based on the calculated trajectory.

[0131] Next, in S43, CPU51 temporarily sets the lane change position (lane change position) on the path of the processing object to be implemented through the branch point of the processing object. Specifically, within the lane change range set in S41 above, the lane change position is temporarily set to the side closest to the destination. For example Figure 18 It means in Figure 17 The diagram shows the temporary lane change positions set on each path. The lane change position is set by a line segment 77 with a length equal to the distance required for the lane change calculated in S42 above. The endpoint on the starting side of line segment 77 corresponds to the lane change start point, and the endpoint on the destination side corresponds to the lane change end point. In S43, the lane change position is temporarily set such that the end point of the lane changeable range is located at the lane change end point.

[0132] Furthermore, as described later, the lane change position temporarily set in S44 above is corrected when the prescribed conditions are met, but if no correction is made, it is basically determined based on the temporarily set lane change position. That is, in this embodiment, the lane change position is preferentially set at a position close to the destination.

[0133] Then, in S44, CPU51 determines whether the lane change implemented to pass through the bifurcation point being processed is a lane change to a lane with a higher lane cost than the current lane. As mentioned above, the lane cost in the overtaking lane is set to a higher value than that of the driving lane. Furthermore, in countries where left-hand traffic is permitted, the value is set higher for lanes further to the right. That is, in the aforementioned S44, for countries where left-hand traffic is permitted, it is determined whether to make a lane change to the right.

[0134] Furthermore, when it is determined that the lane change implemented to pass through the bifurcation point being processed is a lane change to a lane with a higher lane cost than the current lane ("Yes" in S44), the process proceeds to S46. Conversely, when it is determined that the lane change implemented to pass through the bifurcation point being processed is a lane change to a lane with a lower lane cost than the current lane ("No" in S44), the process proceeds to S45.

[0135] In S45, CPU51 corrects the lane change position temporarily set in S43. Specifically, within the lane change range set in S41, it moves the lane change position to the side closest to the origin. The rationale for moving the lane change position towards the origin is that it is appropriate to perform the lane change to a less costly lane as early as possible. Then, after processing other branch points in the path being processed as described in S41, it proceeds to S52.

[0136] On the other hand, in S46, CPU51 obtains the recommended position for lane changing when performing lane changes at the bifurcation point that is the processing target, and uses this as the recommended lane change position. In this embodiment, the lane changes included in the path are based on the premise that the lane changes are performed to pass through the bifurcation point, and the recommended lane change position is set based on the bifurcation point that is the factor in the lane change. For example... Figure 17 or Figure 18 As shown, for lane changes involving moving to the right lane to make a right turn at a fork in the road ahead, the lane change location is set based on the fork in the road where the right turn is intended. Specifically, the recommended lane change location is determined by the type of road entering the fork (i.e., the type of road from which the lane change is taking place). For example, on intercity highways, it is set 2000m before the fork. On urban highways, it is set 1000m before the fork. On other general roads or narrow streets, it is set 700m before the fork.

[0137] Next, in S47, CPU51 determines whether the lane change position temporarily set in S43 is closer to the destination than the lane change recommendation position obtained in S46.

[0138] Furthermore, if it is determined that the lane change position temporarily set in S43 is closer to the destination than the recommended lane change position obtained in S46 ("Yes" in S47), the process moves to S49. Conversely, if it is determined that the lane change position temporarily set in S43 is closer to the origin than the recommended lane change position obtained in S46 ("No" in S47), the process moves to S48.

[0139] In S48, CPU 51 terminates without correcting the lane change position temporarily set in S43. That is, it maintains the lane change position set as close to the destination as possible within the lane change range. The reason for not making a correction is that the currently temporarily set lane change position has become the position for lane changes at a time as close as possible to the recommended lane change position. Then, after processing other branch points included in the path to be processed as described in S41, it moves to S52.

[0140] On the other hand, in S49, CPU51 determines whether the lane change range set in S41 crosses the lane change recommendation position obtained in S46 (i.e., there is a lane change recommendation position within the lane change range).

[0141] Furthermore, if it is determined that the lane change range set in S41 crosses the lane change recommendation position obtained in S46 ("Yes" in S49), the process proceeds to S50. Conversely, if the lane change range set in S41 does not cross the lane change recommendation position obtained in S46, i.e., the starting point of the lane change range is located closer to the destination than the lane change recommendation position ("No" in S49), the process proceeds to S51.

[0142] In S50, CPU51 corrects the lane change position temporarily set in S43. Specifically, it moves the lane change position closer to the recommended lane change position set in S46. More specifically, it moves the lane change start point to a position consistent with the recommended lane change position. The reason for moving the lane change position is that if a lane change can be performed at the recommended lane change position, it is desirable to perform the lane change at the recommended lane change position. Then, after processing other branch points included in the path being processed as described in S41, it moves to S52.

[0143] On the other hand, in S51, CPU51 corrects the lane change position temporarily set in S43. Specifically, within the lane change range set in S41, the lane change position is moved to the side closest to the origin. The reason for moving the lane change position to the origin is to prevent lane changes near forks and to make lane changes as close as possible to the recommended lane change position. Then, after processing other forks in the path being processed as described in S41, the process moves to S52.

[0144] Then, in S52, CPU51, referring to the lane change positions set on the path to be processed, if multiple lane change positions are adjacent, staggers one or both lane change positions and inserts a straight section of a specified distance (e.g., 100m) or more between adjacent lane change positions. Then, the lane change positions set on the path to be processed are determined.

[0145] As a result of the corrections made to the lane change positions in S45, S48, S50, S51, and S52 above, for example, in conjunction with... Figure 18 In the path corresponding to the left image, according to Figure 19 The process shown corrects the lane change position and finally determines it. That is, since the lane change position temporarily set in S43 above is closer to the destination than the recommended lane change position, and the lane change range does not cross the recommended lane change position, the lane change positions of the two lanes are first moved to the side closest to the origin within the lane change range (S51). Then, a straight section of more than a specified distance is inserted between each lane change position and determined (S52).

[0146] Additionally, for example, in relation to Figure 18 In the path corresponding to the middle image, according to Figure 20 The process shown corrects the lane change position and finally determines it. Specifically, since the lane change position corresponding to the first lane change temporarily set in S43 is closer to the destination than the recommended lane change position, and the lane change range extends beyond the recommended lane change position, the lane change position corresponding to the first lane change is first moved to the recommended lane change position (S50). Furthermore, since the lane change position corresponding to the second lane change temporarily set in S43 is closer to the destination than the recommended lane change position, and the lane change range does not extend beyond the recommended lane change position, the lane change position is moved within the lane change range to the side closest to the origin (S51). This state is then determined.

[0147] Additionally, for example, in relation to Figure 18 In the path corresponding to the right image, according to Figure 21 The process shown corrects the lane change position and finally determines it. Specifically, since the lane change position corresponding to the first lane change temporarily set in S43 is closer to the departure point than the recommended lane change position, no correction is made (S48). Furthermore, since the lane change position corresponding to the second lane change temporarily set in S43 is closer to the destination point than the recommended lane change position, and the lane change range does not cross the recommended lane change position, the lane change position is moved within the lane change range to the side closest to the departure point (S51). This state is then determined.

[0148] Then, after performing the processing steps S41 to S52 on the baseline and candidate paths generated in S25 and S26, the process moves to S28. After S28, based on the lane change positions set for the baseline and candidate paths, a recommended lane movement method is selected.

[0149] Next, based on Figure 22 The subprocesses for dynamic driving trajectory generation executed in S6 above will be explained. Figure 22 This is a flowchart of the subprocessing procedure for generating and processing dynamic driving trajectories.

[0150] First, in S61, the CPU 51 acquires the current position of the vehicle detected by the current position detection unit 31. Furthermore, it is preferable to use, for example, high-precision GPS information or high-precision positioning technology to determine the vehicle's current position in detail. High-precision positioning technology refers to a technique that uses image recognition to detect white lines and road surface markings captured by a camera installed on the vehicle, and then matches the detected white lines and road surface markings with, for example, high-precision map information 15, thereby enabling the detection of the driving lane and the high-precision vehicle position. Moreover, when the vehicle is traveling on a road consisting of multiple lanes, the lane in which the vehicle is traveling is also determined.

[0151] Next, in S62, CPU51 acquires the static driving trajectory generated in S3 above (i.e., the predetermined trajectory for the vehicle to travel in the future) and the speed plan generated in S4 above (i.e., the predetermined speed for the vehicle in the future).

[0152] Next, in S63, CPU51, based on the high-precision map information 15 obtained in S2 above, takes the area in front of the vehicle's direction of travel, especially the surrounding area of ​​the "factors that affect the vehicle's driving (hereinafter referred to as influencing factors)" detected in S5 above, as objects, and obtains lane shape, lane line information, etc. Furthermore, the lane shape and lane line information obtained in S63 includes the number of lanes, and information on determining the location and method of adding or removing lanes if there is an increase or decrease in the number of lanes.

[0153] Next, in S64, the CPU51 obtains the current position of the influencing factor detected in S5, as well as the movement status (direction of movement, speed of movement) of the influencing factor if it is moving. Furthermore, the position and movement status of the influencing factor are obtained, for example, by image processing of images captured by the external camera 39 within a defined detection range around the vehicle.

[0154] Then, in S65, CPU51 first predicts the future trajectory of the influencing factor based on the current position and movement status of the influencing factor obtained in S64 above. Additionally, if the influencing factor is another vehicle, the illumination status of the other vehicle's turn signals or brake lights can also be considered for prediction. Furthermore, if the future driving trajectory or speed plan of other vehicles can be obtained through vehicle-to-vehicle communication, these can also be considered for prediction. Then, based on the predicted future trajectory of the influencing factor and the static driving trajectory and speed plan of the vehicle obtained in S62 above, it is determined more accurately whether the influencing factor affects the driving of the vehicle. Specifically, if it is predicted that the vehicle and the influencing factor are in the same lane at the current moment or in the future and the distance between them is close to an appropriate inter-vehicle distance D, it is determined that the influencing factor affects the driving of the vehicle. Furthermore, the appropriate inter-vehicle distance D is calculated, for example, by the following formula (1).

[0155] D = Vehicle speed × 2sec + Vehicle braking distance - Braking distance of influencing factors (However, this is limited to the case where the influencing factor is a moving body)...(1)

[0156] Then, if it is determined that the influencing factor affects the driving of the vehicle ("Yes" in S65), the process moves to S66. On the other hand, if it is determined that the influencing factor does not affect the driving of the vehicle ("No" in S65), the process moves to S9 (S7 and S8 are omitted) without generating a dynamic driving trajectory.

[0157] In S66, CPU51 determines whether a new trajectory (i.e., overtaking) can be generated to allow the vehicle to avoid the influencing factor and return to its static driving trajectory. Specifically, when the influencing factor and the vehicle are in the same lane at the current moment, if the trajectory drawn from the vehicle's ability to overtake the influencing factor by changing lanes to the right within the speed limit and then changing lanes to the left to return to its original lane, and a trajectory maintaining an appropriate distance D or more from the influencing factor is drawn, it is determined that a new trajectory can be generated to allow the vehicle to avoid the influencing factor and return to its static driving trajectory. Furthermore, when the influencing factor and the vehicle are in different lanes at the current moment, and then move into the same lane, if the trajectory drawn from the vehicle's ability to overtake the influencing factor within the speed limit and then changing lanes to the same lane as the influencing factor, and a trajectory maintaining an appropriate distance D or more from the influencing factor is drawn, it is determined that a new trajectory can be generated to allow the vehicle to avoid the influencing factor and return to its static driving trajectory. The determination process in S66 is based on the lane shape and lane line information ahead of the vehicle's direction of travel, the vehicle's current position, the future movement trajectory of influencing factors, and the road's speed limit information obtained in S63.

[0158] Then, if it is determined that a new trajectory can be generated to allow the vehicle to avoid influencing factors and return to the static driving trajectory (i.e., overtaking) ("Yes" in S66), the process proceeds to S67. Conversely, if it is determined that a new trajectory cannot be generated to allow the vehicle to avoid influencing factors and return to the static driving trajectory (i.e., overtaking) ("No" in S66), the process proceeds to S68.

[0159] In S67, CPU51 calculates the trajectory (hereinafter referred to as the avoidance trajectory) used to allow the vehicle to avoid the influencing factor and return to the static driving trajectory (i.e., overtaking). Specifically, when the vehicle and the influencing factor are in the same lane at the current moment, such as Figure 23 As shown, the trajectory of this vehicle—changing lanes to the right to overtake the influencing factor, then changing lanes to the left to return to its original lane—is equivalent to an avoidance trajectory. On the other hand, if the vehicle and the influencing factor are currently in different lanes and need to move into the same lane, such as... Figure 24 As shown, the trajectory of this vehicle after overtaking the influencing factor and then changing lanes to move to the same lane as the influencing factor is equivalent to an avoidance trajectory.

[0160] in, Figure 23 An example is shown of the avoidance trajectory generated in S67 above when the vehicle 90 is traveling in the left lane on a one-way two-lane road and the influencing factor is the vehicle 91 traveling in the same lane ahead.

[0161] First of all, Figure 23 In the example shown, the first trajectory L1 required to begin turning the steering wheel to move into the right lane and then return the steering wheel to the straight-ahead direction is calculated. Furthermore, regarding the first trajectory L1, the lateral acceleration (lateral G) generated during the lane change is calculated based on the vehicle's current speed. A spiral curve is used to calculate a trajectory that is as smooth as possible and minimizes the distance required for the lane change, provided that the lateral G does not exceed an upper limit (e.g., 0.2G) that will not impede autonomous driving support or cause discomfort to the vehicle's occupants. Additionally, maintaining a suitable distance D or more between the vehicle 91 in front is also a condition.

[0162] Next, a second trajectory L2 is calculated, which involves the vehicle in the right lane traveling at the speed limit above the limit, overtaking the vehicle 91 ahead, and maintaining a proper vehicle-to-vehicle distance D above the limit. Furthermore, the second trajectory L2 is essentially a straight line, and its length is calculated based on the speed of the vehicle 91 ahead and the speed limit of the road.

[0163] Next, the third trajectory L3 required to return to the left lane and straight-ahead position by turning the steering wheel is calculated. Furthermore, the third trajectory L3 calculates the lateral acceleration (lateral G) generated during lane change based on the vehicle's current speed. Using a spiral curve as a condition, a trajectory that is as smooth as possible and minimizes the distance required for lane change is calculated, ensuring that the lateral G does not exceed an upper limit (e.g., 0.2G) that will not hinder autonomous driving support or cause discomfort to the vehicle's occupants. Additionally, maintaining a suitable distance D from the vehicle ahead 91 is also a condition.

[0164] in addition, Figure 24 This illustrates an example of the avoidance trajectory generated in S67 above when the vehicle 90 is traveling in the right lane on a one-way two-lane road and needs to move to the left lane, with the influencing factor being the vehicle 91 traveling in front in the left lane.

[0165] First of all, Figure 24 In the example shown, a first trajectory L4 is calculated for a vehicle in the right lane traveling at the speed limit above the vehicle in front, overtaking the vehicle 91 until a suitable vehicle-to-vehicle distance D is achieved. Furthermore, the first trajectory L4 is essentially a straight line, and its length is calculated based on the speed of the vehicle in front 91 and the speed limit of the road.

[0166] Next, the second trajectory L5 required to begin turning the steering wheel to move into the left lane and return the steering wheel position to the straight-ahead direction is calculated. Furthermore, the second trajectory L5 calculates the lateral acceleration (lateral G) generated during lane change based on the vehicle's current speed. Using a spiral curve as a condition, a trajectory that is as smooth as possible and minimizes the distance required for lane change is calculated, ensuring that the lateral G does not exceed an upper limit (e.g., 0.2G) that will not hinder autonomous driving support or cause discomfort to the vehicle's occupants. Additionally, maintaining an appropriate vehicle distance D from the vehicle ahead 91 is also a condition.

[0167] Furthermore, in S67 above, a recommended speed for the vehicle while traveling on the aforementioned avoidance trajectory is calculated. Regarding the recommended speed, using the speed limit as an upper limit, the recommended speed is one where the lateral acceleration (lateral G) generated by the vehicle during lane changes does not exceed an upper limit value (e.g., 0.2G) that does not impede autonomous driving support or cause discomfort to the vehicle's occupants. This is calculated, for example, based on the curvature of the avoidance trajectory and the speed limit.

[0168] Then, in S68, CPU51 calculates the trajectory (hereinafter referred to as the following trajectory) for the vehicle to travel in following (or paralleling) the influencing factor. Specifically, if the vehicle and the influencing factor are in the same lane at the current moment, such as Figure 25 As shown, vehicle 90 continues to travel in its current lane without changing lanes, following the trajectory of an influencing factor (e.g., vehicle 91 ahead), which is equivalent to a following trajectory. Furthermore, the following trajectory is essentially the same as the static driving trajectory. However, since it is necessary to maintain an appropriate inter-vehicle distance from the influencing factor, the speed plan is modified as described later (S8). On the other hand, if the vehicle and the influencing factor are currently in different lanes but need to move to the same lane later, such as... Figure 26 As shown, vehicle 90 continues to travel in its current lane without changing lanes, and the trajectory parallel to influencing factors (such as vehicle 91 ahead) is equivalent to a following trajectory. Furthermore, in this case, the following trajectory becomes a trajectory different from the static driving trajectory.

[0169] Furthermore, in S68 above, a recommended speed for the vehicle while traveling on the aforementioned following trajectory is also calculated. Regarding the following speed of the vehicle, the recommended speed is set at a speed that maintains an appropriate vehicle-to-vehicle distance D or higher from the influencing factors ahead, with the speed limit as the upper limit. The appropriate vehicle-to-vehicle distance D is calculated based on equation (1) above. However, as... Figure 26As shown, it is not necessary to maintain a distance of more than D when the vehicle and the influencing factor are in different lanes. However, considering that the vehicle may need to change lanes to the same lane as the influencing factor later, it is preferable to maintain a distance of more than D.

[0170] Then, in S69, CPU51 generates the avoidance trajectory calculated in S67 above (only if the avoidance trajectory is calculated) and the following trajectory calculated in S68 above, as a dynamic driving trajectory that recommends the vehicle to travel on the roads included in the predetermined driving path based on the surrounding road conditions.

[0171] Then, the dynamic driving trajectory generated in S69 above is stored in flash memory 54, etc., as support information for autonomous driving support.

[0172] Next, based on Figure 27 The subprocesses for the driving trajectory reflection processing performed in S7 above will be explained. Figure 27 This is a flowchart of the sub-processing procedure for reflecting driving trajectory.

[0173] First, in S71, CPU51 reads the static driving trajectory generated in S3 and the dynamic driving trajectory generated in S6 from the storage medium such as flash memory 54.

[0174] Next, in S72, CPU51 calculates the path cost for each driving trajectory read in S71, representing the suitability of the vehicle's driving trajectory. The path cost is calculated considering (a) travel time (average speed), (b) number of lane changes, (c) the location of the lane change, and (d) at least one of the driving lanes. Specifically, the calculation is based on the following conditions.

[0175] (a) Regarding "travel time (average speed)," the longer the travel time (i.e., the slower the average speed), the higher the path cost is calculated. Furthermore, the average speed of a static travel trajectory is determined based on the speed plan generated in S4 above. On the other hand, the dynamic travel trajectory is determined based on the recommended speed calculated in S67 and S68 above.

[0176] (b) Regarding “number of lane changes”, the more lane changes a driving trajectory has, the higher the path cost will be calculated.

[0177] (c) Regarding "the location of lane changes," in the case of multiple lane changes, the shorter the interval between lane changes, the higher the path cost will be calculated. Additionally, for driving trajectories that involve lane changes within a specified distance before an intersection (e.g., 700m for general roads and 2km for highways), the path cost will increase.

[0178] (d) Regarding “driving lanes”, the longer the driving distance in the overtaking lane, the higher the path cost is calculated.

[0179] However, regardless of the conditions in (a) to (d) above, the cost for a driving trajectory determined to be in contact with the influencing factors detected in S5 above will be set to infinite.

[0180] Then, in S73, CPU51 compares the path cost of each driving trajectory calculated in S72 above, and selects the driving trajectory with the smallest path cost as the recommended driving trajectory for the vehicle.

[0181] Next, in S74, CPU51 determines whether an avoidance trajectory or a following trajectory has been selected as the dynamic driving trajectory in the above S73.

[0182] Then, if it is determined that an avoidance trajectory or a following trajectory has been selected as the dynamic driving trajectory in S73 ("yes" in S74), the process proceeds to S75.

[0183] In S75, CPU51 takes the selected dynamic driving trajectory interval as the object and replaces the static driving trajectory with the dynamic driving trajectory as the result. Furthermore, when the static driving trajectory is essentially replaced with a dynamic driving trajectory, the start and end points of the dynamic trajectory are connected to the static driving trajectory, except in exceptional cases where... Figure 26 In the case of the tracking trajectory shown, the end point of the dynamic driving trajectory may not be connected to the static driving trajectory. In such cases, the static driving trajectory can either be regenerated starting from the end point of the dynamic driving trajectory, or the dynamic driving trajectory can be repeatedly generated at certain intervals until it is connected to the static driving trajectory.

[0184] Then, based on a portion of the static driving trajectory that has been replaced with a dynamic driving trajectory, support driving based on autonomous driving support is performed (S9, 10).

[0185] On the other hand, if it is determined that a static driving trajectory was selected in S73 (no in S74), the process proceeds to S8 without replacing the trajectory with a dynamic driving trajectory.

[0186] As detailed above, in the navigation device 1 and the computer program executed by the navigation device 1 in this embodiment, a predetermined driving path for the vehicle is obtained (S1), and using map information including lane shapes, a lane network is used to determine and obtain candidates for lane movement modes that can be selected when the vehicle moves on the predetermined driving path (S25, S26). Then, for the candidates for lane movement modes that accompany lane changes among the obtained lane movement mode candidates, the lane change start position and the lane change end position are preferentially set to positions close to the destination (S43). Since the set lane change start position and lane change end position are taken into account, the recommended lane movement mode for the vehicle to move is selected from the candidates for lane movement modes (S29), so the recommended lane movement mode for lane changes can be appropriately selected. As a result, unrecommended lane changes can be prevented when the vehicle is moving, and driving assistance can be appropriately implemented.

[0187] Furthermore, the present invention is not limited to the above-described embodiments, and various improvements and modifications can be made without departing from the spirit of the present invention.

[0188] For example, in this embodiment, for the candidate lane movement mode accompanied by lane change, the lane change position (77) for lane change is preferably set to a position close to the destination. In subsequent processing, it is shifted towards the starting point to reduce lane change cost. Conversely, the position close to the starting point can also be set as the lane change position (77) for lane change. In subsequent processing, it is shifted towards the destination point on the condition that the lane change cost does not increase.

[0189] In addition, in this embodiment, when factors affecting the vehicle's driving are detected, an avoidance trajectory and a following trajectory are generated as dynamic driving trajectories, but either one can also be generated.

[0190] Furthermore, in this embodiment, the high-precision map information possessed by the server device 4 includes information related to both the road lane shape (road shape, curvature, lane width, etc. in lane units) and the lane lines drawn on the road (lane center line, lane boundary line, lane outer line, guide line, etc.). However, it may only include information related to lane lines or only information related to the road lane shape. For example, even when only information related to lane lines is included, information equivalent to information related to the road lane shape can be inferred based on the information related to lane lines. Furthermore, even when only information related to the road lane shape is included, information equivalent to information related to lane lines can be inferred based on the information related to the road lane shape. Additionally, "information related to lane lines" can be information determining the type and configuration of the lane lines themselves, information determining whether lane changes are possible between adjacent lanes, or information that directly or indirectly determines the shape of the lanes.

[0191] In addition, in this embodiment, a dynamic driving trajectory is generated when factors affecting the vehicle's driving are detected, and the path cost of the existing static driving trajectory is compared with that of the newly generated dynamic driving trajectory (S72, S73). The static driving trajectory is replaced with the dynamic driving trajectory only when it is determined that the dynamic driving trajectory is recommended (S75). However, the static driving trajectory can also be replaced with the dynamic driving trajectory when a dynamic driving trajectory has been generated.

[0192] In addition, in this embodiment, as a unit that reflects the dynamic driving trajectory on the static driving trajectory, a portion of the static driving trajectory is replaced with the dynamic driving trajectory (S75). However, the trajectory can also be corrected so that the static driving trajectory is close to the dynamic driving trajectory, rather than being replaced.

[0193] Furthermore, in this embodiment, as an automatic driving support that is independent of the user's driving operations, all operations related to the vehicle's behavior, namely accelerator operation, brake operation, and steering wheel operation, are described in the case where the vehicle is controlled by the vehicle control ECU 40. However, it is also possible to describe an automatic driving support that includes at least one of the operations related to the vehicle's behavior, namely accelerator operation, brake operation, and steering wheel operation, in the case where the vehicle is controlled by the vehicle control ECU 40. On the other hand, manual driving based on the user's driving operations is described in the case where the user performs all operations related to the vehicle's behavior, namely accelerator operation, brake operation, and steering wheel operation.

[0194] Furthermore, the driving support of this invention is not limited to autonomous driving support for vehicles. For example, in addition to displaying static and dynamic driving trajectories on the navigation screen, driving support can also be provided through guidance using sound, visuals, etc. (e.g., lane change guidance, recommended speed guidance, etc.). Additionally, user driving operations can be supported by displaying static and dynamic driving trajectories on the navigation screen.

[0195] Furthermore, in this embodiment, although the autonomous driving support program ( Figure 4 The operation can be performed by the navigation device 1, but it can also be performed by an in-vehicle unit or vehicle control ECU 40 other than the navigation device 1. In this case, the in-vehicle unit or vehicle control ECU 40 obtains the vehicle's current location, map information, etc., from the navigation device 1 or server device 4. Furthermore, the server device 4 can also execute the autonomous driving support program (…). Figure 4 This refers to one or all of the steps involved. In this case, server device 4 is equivalent to the driving support device of this application.

[0196] In addition to navigation devices, this invention can also be applied to mobile phones, smartphones, tablets, personal computers, etc. (hereinafter referred to as portable terminals, etc.). Furthermore, it can also be applied to systems consisting of servers and portable terminals, etc. In this case, the aforementioned autonomous driving support program (see...) Figure 4 Each step of the process can be implemented using either a server or a portable terminal. However, when applying the invention to a portable terminal or similar device, it is necessary to enable the vehicle capable of performing autonomous driving support to connect to the portable terminal or similar device in a communicable manner (whether wired or wireless).

[0197] Furthermore, the above describes an embodiment of the driving support device of the present invention, but the driving support device can also have the following structure, in which case the following effects can be obtained.

[0198] For example, the first structure is as follows.

[0199] The driving support device includes: a driving predetermined path acquisition unit (51) for acquiring a driving predetermined path for the vehicle (5); a lane network acquisition unit (51) for acquiring a network representing the lane movement that the vehicle can select, i.e., a lane network, based on map information including lane shapes, for the driving predetermined path; a candidate acquisition unit (51) for determining and acquiring candidates for lane movement modes that the vehicle can select when moving on the driving predetermined path through the lane network; a lane change position setting unit (51) for setting the lane change start position and lane change end position of the lane movement mode that accompany the lane change among the candidates for lane movement modes acquired by the candidate acquisition unit, preferably at a position close to the destination; a recommended movement mode selection unit (51) for selecting a recommended lane movement mode for the vehicle when moving from the candidates for lane movement modes, taking into account the lane change start position and lane change end position set by the lane change position setting unit; and a driving support unit (51) for providing driving support to enable the vehicle to move according to the lane movement mode selected by the recommended movement mode selection unit.

[0200] According to the driving support device with the above structure, a candidate lane movement mode that the vehicle can select is determined and obtained through a lane network. For each candidate, the lane change start position and lane change end position are preferentially set to positions close to the destination. In this state, a recommended lane movement mode is selected from the candidate lane movement modes, thus enabling the appropriate selection of the recommended lane movement mode for lane changes. As a result, unrecommended lane changes can be prevented while the vehicle is driving, thereby enabling appropriate driving support.

[0201] In addition, the second structure is described below.

[0202] The lane network is a network that divides the predetermined driving path into multiple lanes and sets nodes (75) for each lane located at the boundary of each divided section, and has road segments (76) connecting the set nodes. For the candidate lane movement mode obtained by the candidate acquisition unit, the lane of the vehicle and the position for lane change are determined by the road segments included in the lane network.

[0203] According to the driving support device with the above structure, it is possible to provide support for driving the vehicle in a lane-moving manner that performs the recommended lane change.

[0204] In addition, the third structure is described below.

[0205] The driving support device includes: a range setting unit (51), which sets a range for lane change that can be performed, i.e. a lane changeable range, for each lane change included in the lane movement mode candidates obtained by the candidate acquisition unit (51) and the lane change position setting unit sets the lane change start position and the lane change end position within the lane changeable range.

[0206] According to the driving support device with the above structure, since the lane change start position and lane change end position are preset to positions that can be implemented for the lane movement mode candidate, driving support can be appropriately implemented based on the selected lane movement mode even if any candidate is selected as the recommended lane movement mode of the vehicle.

[0207] In addition, the fourth structure is described below.

[0208] The recommended movement mode selection unit (51) calculates the cost for each candidate lane movement mode, taking into account the lane change start position and lane change end position set by the lane change position setting unit, and compares the calculated costs to select the recommended lane movement mode for the vehicle when moving from the candidates lane movement modes.

[0209] According to the driving support device with the above structure, by using a lane change start position and lane change end position that take into account costs, it is possible to appropriately select a lane movement method that performs a lane change at the most recommended position. Furthermore, by providing driving support based on the selected lane movement method, driving support can be appropriately implemented.

[0210] In addition, the fifth structure is described below.

[0211] The driving support device includes: a recommended position setting unit (51) for setting a recommended position at a location near the foreground of a lane movement mode candidate that accompanies a lane change for passing a bifurcation point, among the lane movement mode candidates obtained by the candidate acquisition unit (51); and a recommended position correction unit (51) for correcting the lane change start position and lane change end position set by the lane change position setting unit (51) to be closer to the recommended position, among the lane change start position and lane change end position set by the lane change position setting unit (51) to the side of the bifurcation point.

[0212] According to the driving support device with the above structure, the lane change start position and lane change end position set as candidates for lane movement can be corrected to be as close as possible to the recommended lane change position. As a result, driving support can be appropriately implemented based on the lane movement mode that performs a lane change at the recommended position.

[0213] In addition, the sixth structure is described below.

[0214] The recommended movement mode selection unit (51) prioritizes the candidate lane movement mode that changes lanes at a position close to the recommended position as the recommended lane movement mode for the vehicle when moving.

[0215] According to the driving support device with the above structure, a lane movement mode that performs a lane change at a time as close as possible to the recommended lane change position can be selected as the recommended lane movement mode for the vehicle when moving. As a result, unrecommended lane changes can be prevented while the vehicle is moving, thereby enabling appropriate driving support to be implemented.

[0216] In addition, the seventh structure is described below.

[0217] The recommended movement mode selection unit (51) prioritizes the candidate lane movement modes that have a shorter travel distance in the overtaking lane as the recommended lane movement mode for the vehicle when it moves.

[0218] According to the driving support device with the above structure, it is possible to select a lane movement mode that minimizes the distance traveled in the overtaking lane as the recommended lane movement mode for the vehicle when it moves.

[0219] In addition, the eighth structure is described below.

[0220] The driving support device includes: an interval correction unit, which corrects the multiple lane change positions set by the lane change position setting unit (51) to a distance of more than a specified interval for the lane movement mode candidate that is accompanied by multiple lane changes among the lane movement mode candidates obtained by the candidate acquisition unit (51).

[0221] The driving support device with the above structure can correct the lane change position, preventing multiple consecutive lane changes. As a result, it can prevent unrecommended lane changes while the vehicle is driving, thereby enabling appropriate driving support.

[0222] Explanation of reference numerals in the attached figures:

[0223] 1: Navigation device; 2: Driving support system; 3: Information release center; 4: Server device; 5: Vehicle; 15: High-precision map information; 33: Navigation ECU; 39: External camera; 40: Vehicle control ECU; 51: CPU; 52: RAM; 53: ROM; 54: Flash memory; 61: Driving route; 75: Lane node; 76: Lane segment; 77: Lane change location.

Claims

1. A driving support device, wherein a travel scheduled path acquisition unit acquires a travel scheduled path along which a vehicle travels; a lane network acquisition unit acquires, for the travel scheduled path, a network of lane movements that the vehicle can select, based on map information including a lane shape; a candidate acquisition unit determines and acquires, by the network of lane movements, candidates of lane movement manners that the vehicle can select when moving along the travel scheduled path; a lane change position setting unit preferentially sets, for candidates of the lane movement manners that accompany lane changes among the candidates of the lane movement manners acquired by the candidate acquisition unit, a lane change start position at which lane changing starts and a lane change end position at which lane changing ends, at positions close to a destination; a recommended movement manner selection unit selects, from the candidates of the lane movement manners, a lane movement manner of the vehicle that is recommended when the vehicle moves, taking into account the lane change start position and the lane change end position set by the lane change position setting unit; and a driving support unit performs driving support for causing the vehicle to move in the lane movement manner selected by the recommended movement manner selection unit, the candidates of the lane movement manners acquired by the candidate acquisition unit are determined, in units of sections into which the travel scheduled path is divided based on the map information, with respect to a lane along which the vehicle travels and a position at which lane changing is performed, the lane change start position at which lane changing starts and the lane change end position at which lane changing ends are preferentially set, by the lane change position setting unit, at positions close to the destination within a section in which lane changing is determined to be performed.

2. The driving support device according to claim 1, wherein the network of lane movements is a network that divides the travel scheduled path into a plurality of sections and sets nodes with respect to each lane located at a boundary of each section that is divided, and has road segments that connect the set nodes to each other, with respect to the candidates of the lane movement manners acquired by the candidate acquisition unit, a lane along which the vehicle travels and a position at which lane changing is performed are determined by the road segments included in the network of lane movements.

3. The driving support device according to claim 1, wherein the driving support device has a range setting unit that sets, for candidates of the lane movement manners that accompany lane changes among the candidates of the lane movement manners acquired by the candidate acquisition unit, a range in which lane changing can be performed, that is, a lane changeable range, with respect to each lane change included, the lane change position setting unit sets the lane change start position at which lane changing starts and the lane change end position at which lane changing ends within the lane changeable range.

4. The driving support device according to claim 2, wherein the driving support device has a range setting unit that sets, for candidates of the lane movement manners that accompany lane changes among the candidates of the lane movement manners acquired by the candidate acquisition unit, a range in which lane changing can be performed, that is, a lane changeable range, with respect to each lane change included, ​ ​ The lane change position setting unit sets the lane change start position and the lane change end position within the lane changeable range.

5. The driving support apparatus according to any one of claims 1 to 4, wherein The recommended movement manner selection unit calculates a cost in consideration of the lane change start position and the lane change end position set by the lane change position setting unit for each of the candidates of the lane movement manner, The recommended movement manner selection unit compares the calculated costs to select a lane movement manner of the vehicle recommended when the vehicle moves from among the candidates of the lane movement manner.

6. The driving support apparatus according to any one of claims 1 to 4, wherein has: a recommended position setting unit that sets a recommended position at a point near a branch point by a prescribed distance for a candidate of the lane movement manner that accompanies the lane movement in order to pass through the branch point among the candidates of the lane movement manner acquired by the candidate acquisition unit; and a recommended position correction unit that corrects the lane change start position and the lane change end position set on the side closer to the branch point than the recommended position among the lane change start position and the lane change end position set by the lane change position setting unit in a manner to approach the recommended position.

7. The driving support apparatus according to claim 5, wherein has: a recommended position setting unit that sets a recommended position at a point near a branch point by a prescribed distance for a candidate of the lane movement manner that accompanies the lane movement in order to pass through the branch point among the candidates of the lane movement manner acquired by the candidate acquisition unit; and a recommended position correction unit that corrects the lane change start position and the lane change end position set on the side closer to the branch point than the recommended position among the lane change start position and the lane change end position set by the lane change position setting unit in a manner to approach the recommended position.

8. A computer program product including a computer program that causes a computer to function as the following units: a travel scheduled path acquisition unit that acquires a travel scheduled path along which a vehicle travels; a lane network acquisition unit that acquires, based on map information including a lane shape, a network of lane movements, that is, a lane network, that a vehicle can select with respect to the travel scheduled path; a candidate acquisition unit that determines and acquires, by the lane network, candidates of lane movement manners that the vehicle can select when moving on the travel scheduled path; a lane change position setting unit that, for a candidate of the lane movement manner that accompanies a lane change among the candidates of the lane movement manner acquired by the candidate acquisition unit, preferentially sets a lane change start position at which the lane change is started and a lane change end position at which the lane change is ended at a position near a destination; The recommended movement mode selection unit selects a lane movement mode of the vehicle recommended when the vehicle moves, from among the candidates for the lane movement mode, in consideration of the lane change start position and the lane change end position set by the lane change position setting unit. And The driving support unit performs driving support for causing the vehicle to move in the lane movement mode selected by the recommended movement mode selection unit, The candidates for the lane movement mode acquired by the candidate acquisition unit determine the lane in which the vehicle travels and the position at which the lane change is made, in units of sections on the basis of the map information, in which the travel scheduled path is divided into a plurality of sections, The lane change position setting unit preferentially sets, within the section in which the lane change is determined to be made, the lane change start position at which the lane change is started and the lane change end position at which the lane change is ended, at a position close to the destination.

Images