Action plan device, vehicle control system, and action plan generation method

Abstract

To provide an action plan device, a vehicle control system, and an action plan generation method for stopping in an area where impact on other traffic participants is low, without obstructing the progression of the other traffic participants, when abnormality occurs to a detector that recognizes the ambient environment of the host vehicle.SOLUTION: The action plan device comprises: an emergency stoppage request unit that detects abnormality on the basis of obstruct information acquired from an obstruct information acquisition unit and a roadside information acquisition unit and outputs an emergency stoppage request; a stop position determination unit that computes the stop position of the host device on the basis of the emergency stoppage request and determines whether or not being in an area where the progression of other traffic participants is obstructed; and an action plan generation unit that, when it is determined that the host device is going to stop in an area such as an intersection where it hinders the progression of other traffic participants, generates an action plan that encourages stopping outside of the area or passing through the area before stopping.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present application relates to a behavior planning device, a vehicle control system, and a behavior plan generating method. [Background technology]

[0002] In recent years, there has been active development of autonomous driving technology for automobiles, and in addition to providing driving assistance to users, autonomous driving technology without user intervention is attracting attention. When autonomous driving occurs, if an event occurs that makes it impossible to drive safely, such as a failure of some of the sensors that detect the vehicle's surrounding environment, the vehicle is required to stop in a situation where the risk of an accident is low without interfering with other traffic participants.

[0003] For example, in Patent Document 1, when an abnormality is detected in an external environment recognition device that recognizes the surrounding environment of the vehicle, a process is executed to stop the vehicle. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2022-24741 Summary of the Invention [Problem to be solved by the invention]

[0005] In Patent Document 1, when an abnormality occurs in an external recognition system for recognizing the surrounding environment of the vehicle, a trajectory for the vehicle to stop is generated based on MRM (Minimum Risk Maneuver). However, there is a possibility that the vehicle may stop in an area where roads intersect, such as an intersection. If the vehicle stops in an intersection, it may impede the progress of other traffic participants, or there may be a possibility of a collision between the vehicle and the other traffic participants. This is because an intersection is an area where roads intersect, and is an area where the vehicle has a large impact on other traffic participants. In addition, even if the vehicle stops on the shoulder of a road outside the intersection, there is a risk that the vehicle may impede the passage of other traffic participants who are trying to pass through a road connected to the shoulder or a parking lot entrance / exit. These are issues to be addressed when considering the sophistication of automated driving.

[0006] The present application discloses technology for solving the problems described above, and aims to provide a behavior planning device, a vehicle control system, and a behavior plan generation method for stopping a vehicle in an area that does not impede the progress of other traffic participants and has a low impact on those other traffic participants when an abnormality occurs in a detector (sensor) that recognizes the surrounding environment of the vehicle. [Means for solving the problem]

[0007] The behavior planning device disclosed in the present application is Obstacle information around the moving object acquired by the obstacle information acquisition unit; Position information of the moving object acquired by a self-position acquisition unit; Road information around the moving object acquired by a road information acquisition unit, and An action planning device to which obstacle information around a roadside acquired by a roadside information acquisition unit is input, an emergency stop request unit that outputs an emergency stop request when an abnormality is detected in at least one of the obstacle information around the moving object acquired by the obstacle information acquisition unit and the obstacle information around the roadside acquired by the roadside information acquisition unit; a stop position determination unit that determines whether the moving object will stop within a first area that would hinder the progress of other traffic participants when the moving object stops, based on the road information, the position information of the moving object, and the emergency stop request output from the emergency stop request unit; and an action plan generation unit that generates an action plan to avoid the moving body from stopping within the first area when the stop position determination unit determines that the moving body will stop within the first area, which would hinder the progress of other traffic participants. Effect of the Invention

[0008] According to the present application, it is possible to generate an action plan for stopping in an area that does not impede the progress of other traffic participants and has a low impact on other traffic participants. [Brief description of the drawings]

[0009] [Figure 1] 1 is a functional block diagram showing a configuration of a behavior planning apparatus and a vehicle control system according to a first embodiment. [Diagram 2] 4 is a diagram showing an example of an obstacle detection range by an obstacle information detection unit mounted on a vehicle; FIG. [Diagram 3] 4 is a diagram for explaining the operation of a stop position determination unit of the behavior planning device according to the first embodiment. FIG. [Figure 4] 4 is a flowchart showing operations in the behavior planning device and the vehicle control system according to the first embodiment. [Diagram 5] 2 is a functional block diagram showing a configuration of a behavior plan generating unit of the behavior planning device according to the first embodiment. FIG. [Figure 6] 5 is a flowchart showing an operation of a behavior plan generating unit of the behavior planning device according to the first embodiment. [Figure 7] 4 is a diagram for explaining the operation of a behavior plan generating unit of the behavior planning device according to the first embodiment. FIG. [Figure 8] FIG. 2 is a diagram for explaining candidate routes generated by a candidate route generating unit according to the first embodiment. [Figure 9]4 is a diagram for explaining the travel difficulty of a candidate route calculated by a route travel difficulty calculation unit according to the first embodiment. FIG. [Figure 10] 4 is a diagram for explaining an action determined by an action determining unit according to the first embodiment. FIG. [Figure 11] FIG. 11 is a diagram for explaining the operation of a behavior plan generating unit of the behavior planning device according to the second embodiment. [Figure 12] 13 is a diagram for explaining the travel difficulty of a candidate route calculated by a route travel difficulty calculation unit according to the second embodiment. FIG. [Figure 13] FIG. 11 is a diagram for explaining an action determined by a action determining unit according to the second embodiment. [Figure 14] FIG. 11 is a functional block diagram showing a configuration of a behavior planning apparatus and a vehicle control system according to a third embodiment. [Figure 15A] 13 is a flowchart showing an operation of the behavior planning device according to the third embodiment. [Figure 15B] 13 is a flowchart showing an operation of the behavior planning device according to the third embodiment. [Figure 16] FIG. 11 is a diagram for explaining the operation of a behavior plan generating unit of the behavior planning device according to the third embodiment. [Figure 17] FIG. 11 is a diagram for explaining the travel difficulty of a candidate route calculated by a route travel difficulty calculation unit according to the third embodiment. [Figure 18] FIG. 11 is a diagram for explaining an action determined by an action determining unit according to the third embodiment. [Figure 19] 1 is a diagram showing a hardware configuration of a vehicle control system according to first to third embodiments. [Figure 20] FIG. 4 is a diagram illustrating another example of the hardware configuration of the vehicle control system according to the first to third embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, embodiments of the behavior planning device, vehicle control system, and behavior plan generating method disclosed in the present application will be described with reference to the drawings. In the following embodiments, an automobile is taken as an example of a moving body to which the behavior planning device and vehicle control system are applied, and an example of generating a behavior plan for the vehicle is shown. The moving body is assumed to be capable of autonomous driving equivalent to level 3 or level 4 defined by the Society of Automotive Engineers (SAE International), for example. The moving body is assumed to have acquired a driving route to a destination in advance and to drive along the route. In addition, the same reference numerals in each drawing indicate the same or corresponding parts. Therefore, detailed explanations thereof may be omitted to avoid duplication.

[0011] Embodiment 1 The behavior planning device, the vehicle control system, and the behavior plan generating method according to the first embodiment will be described below with reference to the drawings. <Device configuration> FIG. 1 is a functional block diagram showing the configuration of a behavior planning apparatus and a vehicle control system according to the first embodiment. In FIG. 1, the vehicle control system 1000 includes a behavior planning apparatus 100. The behavior planning apparatus 100 is an apparatus for planning the behavior of the host vehicle, and includes an emergency stop request unit 120, a stop position determination unit 140, and a behavior plan generation unit 160 that performs calculations so as to generate behavior that reduces the impact on the surroundings. The vehicle control system 1000 includes an information acquisition unit 300 that acquires information required for generating a behavior plan of the host vehicle in the behavior planning apparatus 100, and the information acquired by the information acquisition unit 300 is input to the behavior planning apparatus 100. The information acquisition unit 300 includes an obstacle information acquisition unit 10 that acquires obstacle information around the host vehicle, a road information acquisition unit 20, a self-position acquisition unit 30 that acquires position information of the host vehicle, and a roadside information acquisition unit 40 that acquires information from a roadside unit RU. Furthermore, the vehicle control system 1000 includes a vehicle control unit 200, which controls each actuator that drives the vehicle, such as the brakes and steering wheel, based on the behavior plan generated by the behavior planning device 100.

[0012] In FIG. 1, the obstacle information acquisition unit 10, the road information acquisition unit 20, the self-position acquisition unit 30, and the roadside information acquisition unit 40 may be included in the behavior planning apparatus 100.

[0013] The obstacle information acquisition unit 10 acquires information about obstacles present around the vehicle detected by the obstacle information detection unit 12 .

[0014] The obstacle information detection unit 12 detects obstacle information. The obstacle is, for example, another vehicle, a pedestrian, a bicycle, a motorcycle, or other traffic participant present around the vehicle. The obstacle information detection unit 12 is, for example, at least one of a camera, a radar, a LiDAR (Light Detection And Ranging), and a sonar sensor provided on the vehicle. The obstacle information detection unit 12 may output the obstacle present around the vehicle to the obstacle information acquisition unit 10 as obstacle information linked to the type of traffic participant classified as another vehicle, a pedestrian, a bicycle, a motorcycle, or the like.

[0015] The camera is installed in a position where it can capture images of the front, sides, and rear of the vehicle, and from the captured images it obtains information indicating the environment in which the vehicle is located, such as information about the lane and obstacles ahead of the vehicle.

[0016] The radar irradiates a radar beam ahead of the vehicle and detects the reflected waves to measure the relative distance and relative speed of an obstacle present ahead of the vehicle, and outputs the measurement results.

[0017] LiDAR detects the position of objects by shining a laser around the vehicle and detecting the time difference between when the laser reflects off nearby objects and when it returns.

[0018] A sonar sensor detects the position and distance of an object by emitting ultrasonic waves to the area around the vehicle and detecting the time difference between when the waves reflect off the surrounding objects and when they return.

[0019] FIG. 2 is a diagram showing the detection range of each obstacle information detection unit 12 in an example in which a plurality of the above-mentioned obstacle information detection units 12 are combined and arranged on a vehicle. In FIG. 2, seven obstacle information detection units 12, from obstacle information detection unit 12a to obstacle information detection unit 12g, are mounted on the host vehicle OV, and three obstacle information detection units 12a, 12b, and 12g are arranged in the front, two obstacle information detection units 12c and 12f are arranged on the sides, and two obstacle information detection units 12d and 12e are arranged in the rear. The detection ranges of each obstacle information detection unit 12 are shown by fan-shaped broken lines, and the host vehicle OV can detect obstacles in the detection ranges SN1, SN2, and SN7 in the front, the detection ranges SN3 and SN6 on the sides, and the detection ranges SN4 and SN5 in the rear. Although not limited to seven obstacle information detection units 12, it is desirable to be able to set the detection range of obstacles so as to cover the periphery of the host vehicle OV.

[0020] The road information acquisition unit 20 acquires information about the roads around the vehicle from a road information detection unit (not shown). The road information acquisition unit 20 includes a map information acquisition unit that has acquired map data of the planned driving route of the vehicle in advance, and acquires road information around the vehicle based on the vehicle position information detected by the vehicle position detection unit 32. Here, the map data includes road information such as the center line of the lane in which the vehicle is traveling, lane width information, stop line information at intersections, information on the number of branches at intersections such as T-junctions and crossroads, stop line information at intersections, and information on the start position of diverging routes. Furthermore, the road information included in the map data may be obtained from the surrounding structure detection results obtained from at least one of a camera, a radar, a LiDAR, and a sonar sensor, which are the obstacle information detection unit 12 provided in the vehicle.

[0021] In order to identify the position of the vehicle, the vehicle is equipped with a GNSS (Global Navigation Satellite System) sensor as a self-position detection unit 32. A GNSS antenna is connected to the GNSS sensor, which receives positioning signals from positioning satellites orbiting in a satellite orbit, analyzes the received positioning signals, and outputs information on the phase center of the GNSS antenna (latitude, longitude, altitude, direction, etc.) to the self-position acquisition unit 30.

[0022] In addition, the vehicle's own position, which is its position information, may be obtained not only by using a GNSS sensor, but also by using SLAM (Simulataneous Localization and Mapping: simultaneous execution of self-position estimation and environmental map creation) technology using surrounding structure detection results obtained from at least one of a camera, radar, LiDAR, and sonar sensor.

[0023] For example, an obstacle information detector (hereinafter referred to as a roadside unit) provided on the roadside includes at least one of a camera, a radar, a LiDAR, and a sonar sensor. The roadside information acquisition unit 40 acquires information about the surroundings of the roadside detected by the roadside unit RU as obstacle information via wireless communication. Note that the roadside unit may be provided not only on the roadside but also on the road such as the roadway, road shoulder, or sidewalk, or may be provided on a building or utility pole near the road. Information about the surroundings of the roadside may be acquired from a remote control system or the like.

[0024] In FIG. 1, the obstacle information acquisition unit 10, the road information acquisition unit 20, the self-position acquisition unit 30, and the roadside information acquisition unit 40 are shown as not being included in the behavior planning device 100. This configuration is suitable for remote control of the vehicle by the control. As an example, the components consisting of the emergency stop request unit 120, the stop position determination unit 140, and the behavior plan generation unit 160 are provided separately from the components consisting of the obstacle information acquisition unit 10, the road information acquisition unit 20, the self-position acquisition unit 30, and the roadside information acquisition unit 40. Specifically, the obstacle information acquisition unit 10, the road information acquisition unit 20, the self-position acquisition unit 30, and the roadside information acquisition unit 40 are provided in the vehicle, and the emergency stop request unit 120, the stop position determination unit 140, and the behavior plan generation unit are provided on the control side as the behavior planning device 100. This is not limited to the above, and may be reversed, for example.

[0025] The emergency stop request unit 120 detects whether or not an abnormality has occurred in the obstacle information acquired from the obstacle information acquisition unit 10 and the obstacle information acquired from the roadside information acquisition unit 40, based on the obstacle information acquired from the obstacle information acquisition unit 10 and the obstacle information acquired from the roadside information acquisition unit 40, and outputs an emergency stop request when an abnormality is detected.

[0026] Here, a method for detecting an abnormality occurring in the obstacle information will be described. First, an example of an abnormality in the obstacle information acquired from the roadside information acquisition unit 40 will be described. An abnormality in the obstacle information acquired from the roadside information acquisition unit 40 is a communication interruption, and the presence or absence of this communication interruption can be determined by providing a signal that counts up at regular intervals from the roadside device to the roadside information acquisition unit 40. When an event is detected in which the count-up at regular intervals does not occur, this can be said to mean that communication from the roadside device to the roadside information acquisition unit 40 has been interrupted. As described above, the emergency stop request unit 120 determines a communication interruption based on the obstacle information acquired from the roadside information acquisition unit 40 and detects an abnormality.

[0027] Next, an example of an abnormality in the obstacle information acquired from the obstacle information detection unit 12 provided in the vehicle will be described. An abnormality in the obstacle information acquired from the obstacle information detection unit 12 provided in the vehicle includes communication interruption and a reduction in the detection range. A communication interruption can be determined and an abnormality can be detected by using a signal that counts up at regular intervals, similar to a communication interruption from a roadside device to a roadside information acquisition unit. The presence or absence of a decrease in the detection range can be determined by checking whether snow or the like has adhered near the sensor, which is the obstacle information detection unit 12, and whether the measured distance continues to be detected as being equal to or less than a predetermined threshold for a predetermined period of time. That is, when snow or the like has adhered so as to cover a portion of the sensor, the detection range is reduced from that in normal times, making it possible to detect an abnormality. Also, when a portion of the imaging element of the camera is damaged, and there is no color in some of the pixels of the camera image, or there is no color change for a predetermined period of time, the detection range is reduced, making it possible to detect an abnormality.

[0028] The stop position determination unit 140 determines, based on the road information acquired from the road information acquisition unit 20, the self-position acquired from the self-position acquisition unit 30, and the emergency stop request output from the emergency stop request unit 120, whether or not to enter and stop in an area that will have a high impact on other traffic participants if the vehicle stops from its current position.

[0029] Here, possible methods for calculating the stopping position include a method in which the current position is calculated as the stopping position as is, or a method in which the stopping position is calculated using the speed (deceleration) that the vehicle can output when stopping from the current vehicle speed.

[0030] Whether or not the calculated stopping position of the vehicle is within an area that has a large impact on other traffic participants is determined based on road information acquired by the road information acquisition unit 20. An area that has a large impact on other traffic participants means an area that would hinder the progress of other traffic participants if the vehicle were to stop at that position, and specifically, is an "area where roads intersect," such as the area inside the stop line at an intersection and the start of a merging point on an expressway. Hereinafter, "an area that has a large impact on other traffic participants" will be read as "an area that would hinder the progress of other traffic participants."

[0031] Next, an example of a method for determining whether or not the host vehicle will stop in an "area that hinders the progress of other traffic participants" will be described with reference to Fig. 3. Fig. 3 is a diagram for explaining the operation of a stop position determination unit of a behavior planning device according to the first embodiment. Fig. 3 shows a state in which the host vehicle OV, which keeps to the left, is approaching an intersection, and the rectangular area indicated by dashed lines connecting the stop lines SL1, SL2, SL3, and SL4 of the lanes that make up the intersection is the intersection area CNF. Here, the inside of the stop lines SL1, SL2, SL3, and SL4 of the intersection, i.e., the area CNF, is defined as the "area that hinders the progress of other traffic participants."

[0032] First, the road information acquisition unit 20 inputs position information of the stop lines SL1, SL2, SL3, and SL4 that form an intersection to the stop position determination unit 140. By connecting the position information of the stop lines SL1, SL2, SL3, and SL4, an area CNF represented by a rectangle can be generated. Then, if the stopping position of the host vehicle OV is inside the area CNF, it is determined that the host vehicle OV will stop in an area that will hinder the progress of other traffic participants.

[0033] Next, the stopping position of the host vehicle OV is calculated. Since center line information of the lane on which the host vehicle is traveling is acquired from the road information acquisition unit 20, if that information is taken as a straight path STR of a target route to be followed by the host vehicle, the stopping position of the host vehicle OV is expressed as a stopping position SP on the straight path STR in Fig. 3. The stopping position SP of the host vehicle OV is calculated using the deceleration of the host vehicle OV on the straight path STR.

[0034] In Figure 3, the position of the center of gravity of the host vehicle OV when it is stopped is represented as the stopping position SP, and if that position is within the intersection area CNF, it is determined that the host vehicle OV will enter an area that will hinder the progress of other traffic participants when stopped. The stop position SP may be any position where the host vehicle OV can be identified, such as the leading or trailing end position of the host vehicle OV. Whether the host vehicle OV is in the area CNF or not may also be determined by generating a rectangle (simulating the external shape of the host vehicle) from the position (stop position) taking into account the size of the host vehicle OV, and if a part of the rectangle is inside the intersection area CNF, it may be determined that the host vehicle OV will enter an area that will hinder the progress of other traffic participants when it stops.

[0035] In the above description, a method of generating an intersection area CNF based on the position information of the stop line of the intersection input from the road information acquisition unit 20 as an area that hinders the progress of other traffic participants has been shown. However, this is not limited to the above. As an area that has a large impact on other traffic participants, an area CNF may be directly input to the stop position determination unit 140 based on intersection information and the like held by the road information acquisition unit 20. Furthermore, by using diverging start position information held by the road information acquisition unit 20, an area including the diverging start position can be used as the area CNF, and an area CNF that is not limited to an intersection area can be set.

[0036] As described above, the stop position determination unit 140 determines whether or not the host vehicle OV will enter and stop in an area that will hinder the progress of other traffic participants when the host vehicle OV stops from its current position.

[0037] Next, the action plan generating unit 160 will be described. The action plan generation unit 160 generates an action plan for avoiding the host vehicle from stopping in an area that would hinder the progress of other traffic participants, based on the obstacle information around the host vehicle output from the obstacle information acquisition unit 10, the road information output from the road information acquisition unit 20, the host vehicle's own position output from the self-position acquisition unit 30, the obstacle information output from the roadside information acquisition unit 40, the emergency stop request output from the emergency stop request unit 120, and the determination result of whether or not to enter and stop in an area that would hinder the progress of other traffic participants output from the stop position determination unit 140.

[0038] The behavior plan of the vehicle generated by the behavior plan generating unit 160 specifically includes a target route, a target speed, and a target position when the vehicle travels. It may also include information indicating upper and lower limits of acceleration when the vehicle travels, and a steering angle for following the target route.

[0039] <Operations of vehicle control system 1000 and behavior planning device 100> Next, the operation of the vehicle control system 1000 and the behavior planning apparatus 100 according to the first embodiment will be described with reference to the flowchart in Fig. 4. The process of the flowchart in Fig. 4 is repeatedly executed while the host vehicle is traveling. Each step in Fig. 4 will be described in association with each functional unit shown in the functional block diagram in Fig. 1.

[0040] First, in step S101, the obstacle information acquisition unit 10 acquires information about obstacles present around the host vehicle, which is output from the obstacle information detection unit 12. In step S 102 , the self-position acquisition unit 30 acquires the self-position output from the self-position detection unit 32 .

[0041] In step S103, the road information acquisition unit 20 acquires road information output from a road information detection unit (not shown) or road information from a map information acquisition unit. In step S104, the roadside information acquisition unit 40 acquires the roadside information output from the roadside unit RU.

[0042] In step S105, the emergency stop request unit 120 detects whether an abnormality has occurred in the obstacle information based on the obstacle information output from the obstacle information acquisition unit 10 and the obstacle information output from the roadside information acquisition unit 40, and outputs an emergency stop request if an abnormality is detected. If an abnormality is detected and a stop request is to be output (Yes in step S105), the process proceeds to step S106. If no abnormality is detected, a stop request is not output (No in step S105), and the process proceeds to step S109.

[0043] When the process proceeds to step S106, the stop position determination unit 140 determines whether the vehicle will stop in an area that would hinder the progress of other traffic participants if the vehicle stops from the current position. If the vehicle will stop in an area that would hinder the progress of other traffic participants (Yes in step S106), the process proceeds to step S107, and if the vehicle will stop without entering an area that would hinder the progress of other traffic participants (No in step S105), the process proceeds to step S108.

[0044] When the process proceeds to step S107, the action plan generating unit 160 generates an action plan that prevents the host vehicle from stopping in an area that would impede the progress of other traffic participants. When the process proceeds to step S108, the action plan generating unit 160 outputs an instruction to stop the vehicle immediately because the stopping position of the vehicle is not an area that will impede the progress of other traffic participants.

[0045] When the process proceeds to step S109, the action plan generating unit 160 outputs a command to continue following the route on which the host vehicle is currently traveling. The output of the action plan generating unit 160 based on steps S107 to S109 is input to the vehicle control unit 200, which controls each actuator (not shown) that drives the vehicle, such as the brakes, steering wheel, etc.

[0046] As described above, when the behavior planning device 100 determines that the vehicle will stop in an area that would hinder the progress of other traffic participants, it generates an action plan to avoid stopping in the area that would hinder the progress of other traffic participants, and when the behavior planning device 100 determines that the vehicle will stop outside the area that would hinder the progress of other traffic participants, it outputs an instruction to stop the vehicle immediately. Also, when no emergency stop request has been generated, it generates an action plan to continue following the currently traveling route.

[0047] <Processing in the Action Plan Generator 160> Next, a process of generating a behavior plan with little impact on the surroundings in the behavior plan generating unit 160 will be described. Fig. 5 is a functional block diagram showing the configuration of the behavior plan generating unit 160. In Fig. 5, the behavior plan generating unit 160 includes a candidate route generating unit 162, a route travel difficulty calculating unit 164, and a behavior deciding unit 166.

[0048] The candidate route generating unit 162 generates routes that the host vehicle can select as candidate routes, based on the road information acquired by the road information acquiring unit 20 and the host vehicle's own position acquired by the host vehicle's own position acquiring unit 30.

[0049] Here, the candidate routes will be explained. In the case of turning right at a crossroads intersection as an example, the road information acquired from the road information acquisition unit 20 is only route information for turning right. However, based on the information that it is a crossroads intersection, the candidate route generation unit 162 also generates a route going straight and a route turning left as a candidate route. In the case of turning right at a T-junction intersection where there is no straight destination as an example, the road information acquired from the road information acquisition unit 20 is only route information for turning right, but based on the information that it is a T-junction intersection, the candidate route generation unit 162 also generates a route turning left as a candidate route.

[0050] The route travel difficulty calculation unit 164 calculates the difficulty of traveling the generated candidate route for each route generated by the candidate route generation unit 162, based on the obstacle information acquired from the obstacle information acquisition unit 10 and the obstacle information output from the roadside information acquisition unit 40. The travel difficulty of the route will be described later.

[0051] The behavior decision unit 166 selects the route with the lowest travel difficulty based on each candidate route and its travel difficulty calculated by the route travel difficulty calculation unit 164, and generates a behavior to avoid stopping in an area that would hinder the progress of other traffic participants. That is, the vehicle is guided to a stopping position set in an area that has little impact on other traffic participants.

[0052] Next, the process of generating a behavior plan with minimal impact on the surroundings in the behavior plan generating unit 160 will be described with reference to the flowchart shown in Fig. 6. The process in the flowchart in Fig. 6 is a detailed version of the operation in step S107 in Fig. 4. Each step in Fig. 7 will be described in association with each functional unit shown in the functional block diagram in Fig. 5.

[0053] First, in step S1071, the candidate route generating unit 162 generates at least one candidate route based on the road information acquired from the road information acquiring unit 20 and the self-position acquired from the self-position acquiring unit 30.

[0054] In step S1072, the route travel difficulty calculation unit 164 calculates a travel difficulty level for each candidate route generated by the candidate route generation unit 162, based on the obstacle information acquired from the obstacle information acquisition unit 10 and the obstacle information acquired from the roadside information acquisition unit 40, when the host vehicle travels along the candidate route.

[0055] In step S1073, the action decision unit 166 selects the route with the lowest travel difficulty level from among the travel difficulty levels of the respective candidate routes calculated by the route travel difficulty level calculation unit 164.

[0056] In step S1074, the behavior decision unit 166 generates an action plan to guide the vehicle based on the selected route to avoid areas that hinder the progress of other traffic participants and to stop in an area that does not hinder the passage of other traffic participants and has a low impact on other traffic participants, and outputs the action plan to the vehicle control unit 200.

[0057] <Operation of the behavior planning device 100> Next, the detailed operation of the behavior planning apparatus 100 will be described with reference to FIGS. FIG. 7 is a diagram for explaining a detailed operation example of the behavior planning apparatus 100, and shows a situation in which an emergency stop request is output from the emergency stop request unit 120 in the host vehicle OV that turns right and passes through an intersection. In FIG. 7, the inside of the stop line of the intersection is expressed as an area CNF that hinders the progress of other traffic participants. In addition, the roadside unit RU can detect obstacles in an area RUA shown in a dotted sector shape, and here, a state in which a pedestrian PD is detected is shown. That is, the roadside information acquisition unit 40 acquires the pedestrian PD as obstacle information. The host vehicle OV acquires obstacle information acquired from the obstacle information detection unit 12, which is an obstacle detection sensor provided in the host vehicle OV, and the roadside unit RU.

[0058] First, in FIG. 7, a situation before an emergency stop request is output from emergency stop request unit 120 will be described. The host vehicle OV travels along the right-turn route RTR, which is acquired from the road information acquisition unit 20 and connects the centers of the roads generated for turning right and passing through the intersection, and based on the position information acquired by the host vehicle position acquisition unit, while being careful of a pedestrian PD, acquired as obstacle information, jumping into the right-turn route RTR.

[0059] It is assumed that the emergency stop request unit 120 detects an abnormality by detecting a failure of the obstacle information detection unit 12 (obstacle detection sensor) that detects the right side of the host vehicle OV with respect to the obstacle information acquired from the obstacle information acquisition unit 10. Then, the emergency stop request unit 120 outputs an emergency stop request to the stop position determination unit 140 and the action plan generation unit 160.

[0060] The stop position determination unit 140 calculates the stop position SP where the host vehicle OV will move to and stop when it stops from its current position based on an emergency stop request. As shown in Fig. 7, the stop position SP is set within an intersection within the area CNF, and if the host vehicle OV stops inside the stop line of the intersection, it will hinder the progress of other vehicles. Then, it is determined that the host vehicle OV will stop within the area CNF, and the determination result is output to the action plan generation unit 160.

[0061] Based on the determination result of the stop position determination unit 140, the vehicle OV generates at least one or more candidate routes in the candidate route generation unit 162 based on the self-position information acquired from the self-position acquisition unit 30 and the road information acquired from the road information acquisition unit 20. Specifically, the exit direction of the intersection is determined from the information corresponding to the type of intersection, which is a crossroad, the information of the stop line SL, and the information of the road width RW acquired from the road information acquisition unit, and a route is generated from the self-position where the vehicle OV is located to the exit direction. FIG. 8 shows an example of the candidate route generated by the candidate route generation unit 162, which shows the candidate route corresponding to the situation of the vehicle in FIG. 7. That is, a straight route STR in the straight direction, a left turn route LTR for turning left, and a right turn route RTR for turning right are generated. Then, the generated candidate route is output to the route travel difficulty calculation unit 164.

[0062] 9 is a diagram for explaining the driving difficulty of each candidate route. The route driving difficulty calculation unit 164 calculates the driving difficulty of the vehicle for each candidate route, i.e., the straight route STR, the left turn route LTR, and the right turn route RTR, generated by the candidate route generation unit 162, using the obstacle information acquired from the obstacle information acquisition unit 10, the self-position information acquired from the self-position acquisition unit 30, and the roadside information acquired from the roadside information acquisition unit 40. A specific calculation method will be described below.

[0063] Since the emergency stop request unit 120 has determined that the obstacle information detection unit 12 that detects the right side of the host vehicle OV has failed, the detectable area on the right side of the host vehicle OV is narrowed. When the obstacle information detection unit 12 mounted on the host vehicle OV is normal, the host vehicle OV runs with increased redundancy by using the obstacle information acquired by the obstacle information acquisition unit 10 and the obstacle information acquired by the roadside information acquisition unit 40. On the other hand, in a situation where the detection area on the right side of the host vehicle OV is narrowed, running on the right turn route RTR that turns right at the intersection means that although the roadside unit RU detects that the pedestrian PD is present near the right turn route RTR, there is a possibility that the response to the pedestrian PD's behavior, for example, the response to the pedestrian jumping out onto the road, may be delayed. Therefore, the difficulty of running on the right turn route RTR is calculated as high.

[0064] Next, for the left turn route LTR, it is determined that there are no obstacles around based on the results of acquisition from the roadside information acquisition unit 40 and the obstacle information acquisition unit 10 at the current position. However, since the plan was to turn right, the steering amount of the steering wheel will be large. Therefore, the driving difficulty level of the left turn route LTR is calculated as medium.

[0065] Regarding the straight route STR, it can be determined that there are no obstacles beyond the area CNF by using the results obtained from the roadside information obtaining unit 40 and the results of the obstacle information detecting unit 12 provided in the host vehicle OV. In addition, since the straight route STR requires less steering of the steering wheel than the left-turn route LTR, the travel difficulty level of the straight route STR is calculated as low.

[0066] The above-mentioned three candidate routes LTR, STR, RTR and the calculation results of the driving difficulty of each are output to the behavior decision unit 166. The behavior decision unit 166 selects the straight route STR with the lowest driving difficulty as the route to be taken by the host vehicle OV, as shown in Fig. 10. Then, the behavior plan for stopping at the stop position MRP after passing through the area CNF that obstructs the progress of other traffic participants on that route is output from the behavior plan generation unit 160.

[0067] In the example of calculating the driving difficulty of each of the candidate routes described above, if an abnormality is detected in the obstacle information detection unit 12 that detects the right side of the host vehicle OV, and the host vehicle was scheduled to travel on the straight-through route STR, the driving difficulty is calculated as follows: the difficulty of traveling on the right-turn route RTR is high, the difficulty of traveling on the straight-through route STR is low, and the difficulty of traveling on the left-turn route LTR is medium. In addition, if an abnormality is detected in the obstacle information detection unit 12 that detects the right side of the host vehicle OV, and the host vehicle was scheduled to travel on the left turn route LTR, the driving difficulty levels are calculated as follows: the difficulty level of traveling on the right turn route RTR is high, the difficulty level of traveling on the straight route STR is medium, and the difficulty level of traveling on the left turn route LTR is low.

[0068] In the above, an example was shown in which an abnormality was detected in the obstacle information detection unit 12 that detects the right side of the host vehicle OV. However, if an abnormality is detected in the obstacle information of the roadside information acquisition unit 40, the behavior of the pedestrian PD on the right side must be detected only by the obstacle information detection unit 12, so a similar driving difficulty level is calculated.

[0069] As described above, according to the first embodiment, the behavior planning device includes an emergency stop request unit, a stop position determination unit, and a behavior plan generation unit, and the emergency stop request unit detects an abnormality based on the obstacle information acquired from the obstacle information acquisition unit and the roadside information acquisition unit, and outputs an emergency stop request when an abnormality is detected, the stop position determination unit calculates the stop position of the host vehicle based on the emergency stop request and determines whether the host vehicle will stop in an area that will hinder the progress of other traffic participants, such as an intersection, and the behavior plan generation unit generates a behavior plan to stop the host vehicle outside the area or after passing through the area, when it is determined that the host vehicle will stop in an area that will hinder the progress of other traffic participants, such as an intersection. Then, the vehicle control unit of the vehicle control system drives each actuator according to the generated behavior plan. This makes it possible to stop the host vehicle in an area that will not hinder the progress of other traffic participants and will have little impact on other traffic participants.

[0070] In addition, the behavior plan generation unit of the behavior planning device is equipped with a candidate route generation unit, a route travel difficulty calculation unit, and an action decision unit, and the candidate route generation unit generates at least one candidate route based on the road information acquired by the road information acquisition unit and the self-position acquired by the self-position acquisition unit, the route travel difficulty calculation unit calculates the travel difficulty when traveling each candidate route based on the obstacle information acquired by the obstacle information acquisition unit and the obstacle information and abnormality detection information acquired by the roadside information acquisition unit, and the action decision unit selects a route with a low travel difficulty, thereby achieving the effect of not only generating a behavior plan to stop after passing through the area, but also selecting a route that has less impact on other traffic participants even in a route that passes through the area.

[0071] Embodiment 2 The behavior planning device, the vehicle control system, and the behavior plan generating method according to the second embodiment will be described below with reference to the drawings. It should be noted that the configurations of the behavior planning apparatus and the vehicle control system according to the second embodiment are similar to those of the first embodiment shown in FIG. 1 and FIG. 5, and therefore a description thereof will be omitted.

[0072] <Operation of the behavior planning device 100> The operation of the behavior planning apparatus 100 according to the second embodiment will be described with reference to FIGS. FIG. 11 is a diagram for explaining a detailed operation example of the behavior planning apparatus 100 according to the second embodiment, and shows a state in which the host vehicle OV turns left and passes through an intersection by following the left-turn route LTR acquired from the road information acquisition unit 20 using the obstacle information detection unit 12 provided in the host vehicle OV and the obstacle information acquired from the roadside unit RU. As in the first embodiment, the inside of the stop line of the intersection is expressed as an area CNF that obstructs the progress of other traffic participants. However, the difference is that the road width in the direction in which the host vehicle OV turns right or left is wider than the road width in the direction in which the host vehicle OV travels straight. In addition, the roadside unit RU can detect obstacles in the area RUA, and in this example, it shows that the state in which the pedestrians PD1 and PD2 are about to cross the road at a position beyond the intersection as viewed from the host vehicle OV, and the state in which the other vehicle VE is stopped by the roadside at a position where the pedestrians have turned left and passed the intersection is detected. That is, the roadside information acquisition unit 40 acquires the pedestrians PD1 and PD2 and the other vehicle VE (stationary obstacle) as obstacle information.

[0073] The host vehicle OV not only follows the left-turn route LTR, which is acquired from a road information acquisition unit 20 and connects the centers of the roads generated for turning left and passing through an intersection, but also avoids the other vehicle VE, which is a stationary obstacle, based on the position information acquired by the host vehicle position acquisition unit 30.

[0074] It is assumed that the emergency stop request unit 120 judges an abnormality in the obstacle information acquired from the obstacle information acquisition unit 10 based on the detection result that an obstacle detection sensor detecting the right side of the host vehicle OV has failed. Then, the emergency stop request unit 120 outputs an emergency stop request to the stop position determination unit 140 and the action plan generation unit 160.

[0075] The stop position determination unit 140 calculates the stop position SP where the host vehicle OV will stop when it stops from its current position based on the emergency stop request, in the same manner as in embodiment 1. Then, as shown in FIG. 11, if the stop position SP is within the area CNF, it determines that the host vehicle OV will stop within the area CNF, and outputs the determination result to the action plan generation unit 160.

[0076] Based on the determination result of the stop position determination unit 140, the host vehicle OV generates at least one or more candidate routes in a candidate route generation unit 162 based on the self-position information acquired from the self-position acquisition unit 30 and the road information acquired from the road information acquisition unit 20. The method of generating the candidate routes is the same as in the first embodiment, and each of a straight route STR, a left turn route LTR, and a right turn route RTR is generated and output to a route travel difficulty calculation unit 164.

[0077] FIG. 12 is a diagram for explaining the travel difficulty of each candidate route. The route travel difficulty calculation unit 164 calculates the travel difficulty for each candidate route, i.e., the straight route STR, the left turn route LTR, and the right turn route RTR, using the obstacle information acquired from the obstacle information acquisition unit 10, the self-position information acquired from the self-position acquisition unit 30, and the roadside information acquired from the roadside information acquisition unit 40, for the routes generated by the candidate route generation unit 162. In addition, road information acquired from the road information acquisition unit 20 is also used as necessary. A specific calculation method will be described below.

[0078] The emergency stop request unit 120 determines that the obstacle detection sensor installed on the right side of the vehicle OV has failed, and therefore the detectable area on the right side of the vehicle OV is narrowed. In a situation where the detection area is narrowed, the travel difficulty of the route traveling in the corresponding direction, i.e., the right turn route RTR, is calculated as high.

[0079] Next, for the straight route STR, the roadside unit RU detects that pedestrians PD1 and PD2 are crossing the road at the point where the vehicle has gone straight through the intersection, so the driving difficulty is calculated as medium, since it is necessary to be careful of the presence of pedestrians PD1 and PD2 after passing through the intersection.

[0080] Regarding the left-turn route LTR, the roadside unit RU detects the presence of another vehicle VE stopped at the side of the road as an obstacle, and therefore calculates the remaining route width LFW for the width of the left-turn route LTR by subtracting the width information of the other vehicle VE present on the route from the road width RW2. FIG. 12 illustrates a situation in which the width of the left-turn route LTR of the host vehicle OV is wider than the road width RW1 of the straight route STR in the forward direction, and it is determined that the remaining route width LFW is greater than the width of the host vehicle OV by a predetermined value or more. Here, the fact that the remaining route width LFW is greater than the width of the host vehicle OV by a predetermined value or more means that there is no need to travel beyond another lane, and the influence of obstacles in another lane can be reduced, so the difficulty of traveling the left-turn route LTR is also calculated as medium.

[0081] The above-mentioned three candidate routes LTR, STR, RTR and the calculation results of the driving difficulty of each are output to the behavior decision unit 166. The behavior decision unit 166 determines the candidate route calculated to have the lowest driving difficulty as the route to be taken by the host vehicle OV. Here, the straight route STR and the left turn route LTR are calculated to have the same driving difficulty, and if they are calculated to have the same driving difficulty, it is determined that the left turn route LTR acquired from the road information acquisition unit 20, i.e., the initially planned left turn route LTR, is to be given priority.

[0082] That is, as shown in Fig. 13, the left turn route LTR is determined as the route to be taken by the host vehicle OV. Then, the action plan generating unit 160 generates an action plan for stopping the host vehicle OV at a stop position MRP after passing through the area CNF on the route. The stop position MRP may be determined to be a position where the host vehicle OV stops after passing through another vehicle VE, which is a stationary obstacle on the left side of the host vehicle OV.

[0083] In this embodiment, the method of giving priority to the route acquired from the road information acquisition unit 20 has been described, but the present invention is not limited to this example. A method of calculating the travel difficulty level in advance from only the candidate routes generated by the candidate route generation unit 162 may also be considered. For example, a method of determining the travel difficulty level of the straight route STR as low, the travel difficulty level of the left turn route LTR as medium, and the travel difficulty level of the right turn route RTR as high in advance, and determining the travel route as the straight route STR in the action decision unit 166 only according to the ranking may also be considered. In addition, as described in the second embodiment, when the travel difficulty levels of the straight route STR and the left turn route LTR are the same, the method of determining the straight route STR by superimposing the predetermined ranking and the travel difficulty level calculated by the route travel difficulty calculation unit may also be considered.

[0084] As described above, according to the second embodiment, the same effect as that of the first embodiment can be achieved. That is, when an abnormality is detected based on the obstacle information, it is possible to stop the vehicle in an area where the vehicle does not impede the progress of other traffic participants and has a small impact on the other traffic participants. Furthermore, when multiple calculation results from the route driving difficulty calculation unit are identical, priority is given to the one that is consistent with the route acquired from the road information acquisition unit, thereby minimizing the difficulty of driving while at the same time enabling the vehicle to perform actions to follow the route set as its target.

[0085] Embodiment 3 The behavior planning device, the vehicle control system, and the behavior plan generating method according to the third embodiment will be described below with reference to the drawings. <Device configuration> Fig. 14 is a functional block diagram showing the configuration of a behavior planning apparatus and a vehicle control system according to the third embodiment. In Fig. 14, a vehicle control system 1000 includes a behavior planning apparatus 100. Note that descriptions overlapping with those of the first and second embodiments will be omitted. As shown in Fig. 14, the behavior planning apparatus 100 further includes an information accumulation unit 180 in addition to the configuration of the behavior planning apparatus 100 shown in Fig. 1, and the behavior plan generated in the behavior plan generation unit 160 includes obstacle information acquired by a roadside information acquisition unit and accumulated in the information accumulation unit 180, and generates a behavior plan for avoiding the host vehicle from stopping in an area that hinders the progress of other traffic participants.

[0086] The information accumulation unit 180 continues to accumulate the obstacle information acquired by the roadside information acquisition unit 40, and outputs the accumulated information to the behavior plan generation unit 160. The obstacle information accumulated in the information accumulation unit 180 makes it possible to know the transition of the obstacle information from the past. As a method of outputting to the behavior plan generation unit 160, a method of outputting only the obstacle information from the time one cycle before the current time t, or a method of outputting information measured over T seconds together with the measurement time, etc., are conceivable. Here, one cycle is the above-mentioned count-up cycle.

[0087] The action plan generating unit 160 includes a candidate route generating unit 162, a route travel difficulty level calculating unit 164, and an action deciding unit 166, as in the first and second embodiments (see FIG. 5). The candidate route generating unit 162 generates candidate routes that can be selected by the host vehicle, based on the road information acquired by the road information acquiring unit 20 .

[0088] The route travel difficulty calculation unit 164 calculates the travel difficulty level for each candidate route generated by the candidate route generation unit 162, based on the obstacle information acquired from the obstacle information acquisition unit 10, the obstacle information output from the roadside information acquisition unit 40, and the obstacle information accumulated in the information accumulation unit 180.

[0089] The behavior decision unit 166 selects the candidate route with the lowest driving difficulty based on each candidate route and its driving difficulty calculated by the route driving difficulty calculation unit 164, and determines an action plan that avoids stopping in areas that would hinder the progress of other traffic participants.

[0090] <Operations of vehicle control system 1000 and behavior planning device 100> 15A and 15B are flowcharts showing the overall operation of the vehicle control system 1000 and the action planning apparatus 100 in the embodiment 3. Note that the processing of the flowcharts in Fig. 15A and 15B is repeatedly executed while the host vehicle is traveling.

[0091] Each step in Figures 15A and 15B will be described below in association with each functional unit shown in the functional block diagram of Figure 14. Note that the operations from step S301 to step S304, step S306, step S3071 to step S3074, step S308, and step S309 are similar to the operations from step S101 to step S104, step S105, step S1071 to step S1074, step S108, and step S109 shown in Figures 4 and 5 of the first embodiment.

[0092] First, in steps S301 to S304, similarly to the first embodiment, each functional unit acquires each piece of information. In step S305, the information accumulation unit 180 accumulates the obstacle information acquired by the roadside information acquisition unit 40.

[0093] In step S306, if an abnormality is detected in the obstacle information, an emergency stop request is output (Yes in step S306), and in step S307, the stop position determination unit 140 determines whether or not the vehicle will stop in an area that will hinder the progress of other traffic participants. In step S307, if it is determined that the vehicle will stop in an area that will hinder the progress of other traffic participants (Yes in step S307), the process proceeds to step S3071.

[0094] In step S3071, the candidate route generating unit 162 generates at least one candidate route based on the road information acquired from the road information acquiring unit 20 and the self-position acquired from the self-position acquiring unit 30.

[0095] In step S3072, for each of the candidate routes generated in step S3071, the route travel difficulty calculation unit 164 calculates a travel difficulty level for the host vehicle to travel on the basis of the obstacle information acquired from the obstacle information acquisition unit 10, the obstacle information acquired from the roadside information acquisition unit 40, and the obstacle information accumulated in the information accumulation unit 180.

[0096] In step S3073, the action decision unit 166 selects the route with the lowest travel difficulty level from among the travel difficulty levels of the respective candidate routes calculated by the route travel difficulty level calculation unit 164.

[0097] In step S3074, the behavior decision unit 166 generates an action plan based on the selected route to guide the vehicle to avoid areas that hinder the progress of other traffic participants and to stop outside those areas or in an area that has a low impact on other traffic participants, and outputs the action plan to the vehicle control unit 200.

[0098] <Operation of the behavior planning device 100> Next, a detailed operation of the behavior planning apparatus 100 according to the third embodiment will be described with reference to FIGS. FIG. 16 shows how the host vehicle OV passes through an intersection by following the straight route STR acquired from the road information acquisition unit 20 using the obstacle information detection unit 12, which is an obstacle detection sensor provided in the host vehicle OV, and obstacle information acquired from the roadside unit RU. As in the first and second embodiments, the inside of the stop line of the intersection is expressed as an area CNF that has a large impact on other traffic participants. The roadside unit RU can detect obstacles in the area RUA, and here shows a detection situation in which the other vehicle VE stops on the road at a position where the other vehicle VE has passed through the intersection by going straight. That is, the other vehicle VE is acquired as obstacle information as a stationary obstacle in the roadside information acquisition unit 40.

[0099] The host vehicle OV not only follows the straight route STR, which is acquired from the road information acquisition unit 20 and connects the center of the road, and is generated to pass through the intersection by going straight, but also avoids the other vehicle VE, which is a stationary obstacle, based on the position information acquired by the host vehicle position acquisition unit 30.

[0100] It is assumed that the emergency stop request unit 120 judges an abnormality in the obstacle information acquired from the roadside information acquisition unit 40 based on the detection result that communication from the roadside unit RU has been interrupted. Then, the emergency stop request unit 120 outputs an emergency stop request to the stop position determination unit 140 and the action plan generation unit 160.

[0101] The stop position determination unit 140 calculates the stop position SP where the host vehicle OV will move and stop when it stops from its current position based on an emergency stop request, in the same manner as in the first and second embodiments. As shown in Fig. 16, when the stop position SP is within the area CNF, it determines that the host vehicle OV will enter the area CNF and stop, and outputs the determination result to the action plan generation unit 160.

[0102] Based on the determination result of the stop position determination unit 140, the vehicle OV generates at least one or more candidate routes in a candidate route generation unit 162 based on the self-position information acquired from the self-position acquisition unit 30 and the road information acquired from the road information acquisition unit 20. The method of generating the candidate routes is the same as in the first and second embodiments. Since the road is a T-junction in this case, a straight route STR and a left-turn route LTR are generated as candidate routes and output to the route travel difficulty calculation unit 164.

[0103] 17 is a diagram for explaining the travel difficulty of each candidate route. The route travel difficulty calculation unit 164 calculates the travel difficulty for each candidate route generated by the candidate route generation unit, that is, the straight route STR and the left turn route LTR, using the obstacle information acquired from the obstacle information acquisition unit 10, the self-position information acquired from the self-position acquisition unit 30, the roadside information acquired from the roadside information acquisition unit 40, and the obstacle information accumulated in the information accumulation unit 180.

[0104] Here, the route travel difficulty calculation unit 164 calculates the travel difficulty for each candidate route, but at the time when communication with the roadside unit RU is interrupted and an abnormality is determined, the obstacle information acquired from the roadside information acquisition unit 40 cannot be used. Therefore, the obstacle information obtained from the roadside unit RU uses the obstacle information previously acquired by the information accumulation unit 180, i.e., the transition of the obstacle information accumulated before the abnormality determination. According to this information, based on the obstacle information previously acquired by the information accumulation unit 180, it is known that another vehicle VE is present at a position where the vehicle has passed through the intersection by going straight ahead. Since the obstacle information previously acquired is of a stationary obstacle, the route travel difficulty calculation unit 164 determines that the obstacle has not moved significantly from that position even now, and estimates that a stationary obstacle is present on the straight route STR.

[0105] As in the second embodiment, the remaining path width LFW relative to the width of the straight path STR is calculated by subtracting width information of the other vehicle VE, which is an obstacle present in the path, from the straight path STR. In the situation of FIG. 17, the remaining path width LFW is determined to be less than the width of the host vehicle OV, and in order to avoid the other vehicle VE, which is an obstacle, the host vehicle needs to travel by protruding into another lane OL (here, the oncoming lane). However, when traveling by protruding into the other lane OL, it is necessary to take into account obstacle information on the other lane OL, but since information from the roadside unit RU has been cut off, it is unclear whether another vehicle will appear on the other lane OL, and if another vehicle does appear, it may hinder travel.

[0106] In addition, when it is assumed that the host vehicle OV stops after traveling to the other vehicle VE, which is an obstacle, it is necessary to consider the size of the area FSP sandwiched between the other vehicle VE, which is an obstacle, and the area CNF. If the area FSP does not include an obstacle but is not longer than the length of the host vehicle OV by a predetermined value or more, the area CNF will be included in the stopping position of the host vehicle OV. In the situation of FIG. 17, since the length of the area FSP is not longer than the length of the host vehicle OV, if the area occupied by the host vehicle OV when traveling on the straight path STR and stopping is the area ESP shown in a thick frame, it is estimated that the host vehicle OV will stop in a situation where a part of the area ESP is included in the area CNF. Therefore, the traveling difficulty of the straight path STR is calculated to be high. In this way, when an obstacle exists beyond the area CNF, and at least a part of the area FSP required when the host vehicle OV stops in front of the obstacle exists within the area CNF, the traveling difficulty is calculated to be high.

[0107] On the other hand, with regard to the left turn route LTR, it is possible to determine that there are no obstacles in the vicinity using information from the obstacle information detection unit 12, which is an obstacle detection sensor provided in the vehicle OV, so that the travel difficulty level of the route in the left turn direction is calculated to be low.

[0108] The above-mentioned two directions of candidate routes STR and LTR and the calculation results of the driving difficulty of each are output to the behavior decision unit 166. The behavior decision unit 166 decides the left turn route LTR, which has the lowest driving difficulty level, as the route to be taken by the host vehicle OV, as shown in Fig. 18. Then, the behavior plan for stopping at the stop position MRP after passing through the area CNF on that route is output from the behavior plan generation unit 160.

[0109] As described above, according to the third embodiment, the same effects as those of the first embodiment can be achieved. That is, when an abnormality is detected based on the obstacle information, it is possible to stop the vehicle in an area where the vehicle does not impede the progress of other traffic participants and has a small impact on the other traffic participants. Furthermore, since the behavior planning device according to the third embodiment further includes an information storage unit, when an abnormality is detected in the obstacle information acquired from the roadside information acquisition unit, the state of the obstacle can be estimated based on the information accumulated up until the occurrence of the abnormality. By using this to calculate the difficulty level of the travel route, it becomes possible to obtain a more accurate difficulty level.

[0110] In particular, the obstacle information around the vehicle acquired by the obstacle information acquisition unit changes the range of detectable obstacle information as the vehicle approaches an area such as an intersection that has a large impact on other traffic participants. However, since the obstacle information acquired by the roadside information acquisition unit is obstacle information from an obstacle detector installed on the side of the road, it is possible to obtain obstacle information within a predetermined range regardless of the position of the vehicle. Therefore, by storing the obstacle information acquired by the roadside information acquisition unit, even if communication is interrupted, it is considered that the position of a stationary obstacle will not change significantly, so it is useful to use the stored past information when calculating the driving difficulty of each candidate route. As a result, even if the detection distance of the obstacle information detection unit (obstacle detection sensor) provided in the vehicle is short, it is possible to determine a route with low driving difficulty by supplementing distant information with roadside information, and to avoid stopping the vehicle in an area that hinders the progress of other traffic participants and to guide the vehicle to a position with low impact on other traffic participants.

[0111] Each functional unit of the behavior planning apparatus 100 in the above-mentioned first to third embodiments may be realized as a hardware configuration of a vehicle control system 1000 illustrated in Fig. 19, that is, a vehicle control system 1000 including a calculation processing circuit 1001, a storage device 1002 including a ROM (Read Only Memory) in which a program for executing the function of each functional unit is stored and a RAM (Random Access Memory) for saving each data of the execution result of each functional unit which is a calculation result by the program, and an input / output circuit 1003. The input / output circuit 1003 receives as inputs an output result on obstacle information from an obstacle information detection unit 12 which is an obstacle detection sensor provided in the vehicle, self-position information detected by a self-position detection unit 32, and an output result on obstacle information from a roadside unit RU, and the like, and outputs control signals such as the speed and handling of the vehicle from the input / output circuit 1003 to a drive system device 2000 which is each actuator for driving the vehicle such as a brake and a handle.

[0112] A processor such as a CPU (Central Processing Unit) or a DSP (Digital Signal Processor) is applied to the arithmetic processing circuit 1001. Dedicated hardware may also be applied to the arithmetic processing circuit 1001. When the arithmetic processing circuit 1001 is dedicated hardware, the arithmetic processing circuit 1001 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination of these.

[0113] Furthermore, the functional units of the behavior planning apparatus 100 and the vehicle control system 1000 may be realized by individual arithmetic processing circuits, or may be collectively realized by a single arithmetic processing circuit. Furthermore, each functional unit of the behavior planning device 100 and the vehicle control system 1000 can realize the above-mentioned functions by hardware, software, or a combination of these, with some functions being realized by a calculation processing circuit as dedicated hardware and other functions being realized by software.

[0114] 20 shows another example of the hardware configuration of the vehicle control system 1000. A communication circuit 1004 is further provided in addition to the configuration in FIG. The communication circuit 1004 includes a wide-area communication unit and a narrow-area communication unit as a communication module. The wide-area communication unit uses a predetermined wide-area wireless communication standard, such as LTE (Long Term Evolution) or 4G or 5G (5th Generation; 5th generation mobile communication system). The narrow-area communication unit uses, for example, DSRC (Dedicated Short Range Communications), and although not described in the above embodiment, it can be used for communication with other vehicles to obtain information about other vehicles around the vehicle. A certain communication speed is guaranteed for these communications.

[0115] Since the roadside unit RU is installed outside the vehicle or outside of control, the obstacle information from the roadside unit RU is received through communication with the vehicle control system 1000. Here, the communication uses, for example, LTE or 5G.

[0116] Each sensor of the obstacle information detection unit 12 and the GNSS sensor of the self-position detection unit 32 are mounted on the vehicle, and information is output from each sensor to the information acquisition unit 300 via a communication line, and is connected using, for example, a CAN (Control Area Network: registered trademark) or the like.

[0117] Furthermore, even when a part of the vehicle control system 1000 is outside the vehicle itself, such as a traffic control station, and activates an emergency stop, signals are sent and received via the communication circuit 1004 .

[0118] <Other embodiments> In the above, the behavior planning apparatus 100 and the vehicle control system 1000 are applied to an automobile as an example of a moving body, but the application is not limited to an automobile, and the behavior planning apparatus 100 can be applied to various other moving bodies. The behavior planning apparatus 100 can be used as an apparatus for planning the behavior of a moving body such as an in-building mobile robot that inspects the inside of a building, a line inspection robot, and a personal mobility. When the moving body is other than an automobile, it is also possible to acquire the drivable area of ​​the route on which the moving body travels as the road information in the detection information in the road information detection unit. As the obstacle information acquired by the roadside unit RU, for example, information from an obstacle information detection unit provided in a building, a line, or within the range in which the personal mobility moves may be used.

[0119] Although the present application describes various exemplary embodiments and examples, the various features, aspects, and functions described in one or more embodiments are not limited to application to a particular embodiment, but may be applied to the embodiments alone or in various combinations. Therefore, countless modifications not exemplified are assumed within the scope of the technology disclosed in the present specification, including, for example, modifying, adding, or omitting at least one component, and further, extracting at least one component and combining it with a component of another embodiment.

[0120] Various aspects of the present disclosure are summarized below as appendices.

[0121] (Appendix 1) Obstacle information around the moving object acquired by the obstacle information acquisition unit; Position information of the moving object acquired by a self-position acquisition unit; Road information around the moving object acquired by a road information acquisition unit, and An action planning device to which obstacle information around a roadside acquired by a roadside information acquisition unit is input, an emergency stop request unit that outputs an emergency stop request when an abnormality is detected in at least one of the obstacle information around the moving object acquired by the obstacle information acquisition unit and the obstacle information around the roadside acquired by the roadside information acquisition unit; a stop position determination unit that determines whether the moving object will stop within a first area that would hinder the progress of other traffic participants when the moving object stops, based on the road information, the position information of the moving object, and the emergency stop request output from the emergency stop request unit; and a behavior plan generation unit that generates a behavior plan for avoiding the moving object from stopping within the first area when the stop position determination unit determines that the moving object will stop within the first area where it will hinder the progress of other traffic participants. (Appendix 2) The action plan generation unit a candidate route generating unit that generates at least one candidate route along which the moving object will travel, based on the road information and the self-position of the moving object; a route travel difficulty calculation unit that calculates a travel difficulty level for the moving body when traveling along each of the candidate routes generated by the candidate route generation unit, based on the obstacle information acquired by the obstacle information acquisition unit, the obstacle information around the roadside acquired by the roadside information acquisition unit, the self-position of the moving body, and the abnormality information detected by the emergency stop request unit; and a behavior determination unit that selects a travel route with the smallest travel difficulty for each of the candidate routes calculated by the route travel difficulty calculation unit, 2. The behavior planning device according to claim 1, further comprising: a behavior planning device that generates a behavior plan for the moving body to stop outside the first region. (Appendix 3) The route travel difficulty calculation unit is If an obstacle exists beyond the first area, a second area is calculated as a stopping area required for the moving body to stop in front of the obstacle; 3. The action planning device according to claim 2, wherein the travel difficulty level is calculated to be high when at least a portion of the second area exists within the first area. (Appendix 4) The candidate route generation unit A route along which the moving object is to travel to a target position is generated as one of candidate routes; The action determination unit is 4. The behavior planning device according to claim 2 or 3, wherein, when the travel difficulty levels are the same, a route along which the moving object is scheduled to move to a target position is preferentially selected. (Appendix 5) An information storage unit that stores obstacle information around the roadside acquired by the roadside information acquisition unit, The action plan generation unit 5. The behavior planning device according to claim 1, wherein, when the stop position determination unit determines that the vehicle will stop in the first area where it will hinder the progress of other traffic participants, the obstacle information accumulated in the information accumulation unit is added to generate a behavior plan for avoiding the vehicle stopping in the first area. (Appendix 6) An information storage unit that stores obstacle information around the roadside acquired by the roadside information acquisition unit, The route travel difficulty calculation unit is 5. The behavior planning device according to claim 2, further comprising: a candidate route generating unit that generates a candidate route for each of the candidate routes along which the moving object travels; (Appendix 7) The action plan generation unit The action planning device described in Appendix 5, wherein, when the emergency stop request unit detects an abnormality in the obstacle information around the roadside acquired by the roadside information acquisition unit, the emergency stop request unit generates an action plan to avoid stopping within the first area using the obstacle information accumulated in the information accumulation unit, without using the obstacle information around the roadside acquired by the roadside information acquisition unit. (Appendix 8) 8. The action planning device according to claim 1, further comprising: the obstacle information acquisition unit, the self-position acquisition unit, the road information acquisition unit, and the roadside information acquisition unit. (Appendix 9) A vehicle control system comprising: the behavior planning device according to any one of claims 1 to 8; and a vehicle control unit that controls a moving object. (Appendix 10) A behavior plan generating method executed by using a behavior planning device, comprising: acquiring obstacle information around the moving object; acquiring location information of the moving object; acquiring road information around the moving object; Obtaining obstacle information around a roadside; outputting an emergency stop request when an abnormality is detected in at least one of the acquired obstacle information around the moving object and the acquired obstacle information around the roadside; determining whether the moving body will stop in an area that would hinder the progress of other traffic participants when stopping the moving body, based on the acquired road information and position information of the moving body and the output emergency stop request; When it is determined that the moving body will stop within the area, generating an action plan for avoiding the moving body from stopping within the area. (Appendix 11) In the step of generating an action plan, generating at least one candidate route along which the moving object will travel based on the road information and the position information of the moving object; calculating a travel difficulty level for the moving body when the moving body travels along each of the candidate routes based on obstacle information around the moving body, obstacle information around the roadside, position information of the moving body, and detected abnormality information; selecting a travel route with the lowest travel difficulty level from among the calculated travel difficulty levels; A method for generating an action plan according to claim 10, comprising: generating an action plan for the moving body to stop outside the area based on a selected travel route. (Appendix 12) After the step of acquiring obstacle information around the roadside, The method further includes a step of accumulating the obtained obstacle information around the roadside, In the step of generating an action plan, 12. The method for generating a behavior plan according to claim 10 or 11, further comprising: generating a behavior plan for the moving object by adding the accumulated obstacle information about the roadside surroundings. [Explanation of symbols]

[0122] 10: Obstacle information acquisition unit, 12, 12a, 12b, 12c, 12d, 12e, 12f, 12g: Obstacle information detection unit, 20: Road information acquisition unit, 30: Self-position acquisition unit, 32: Self-position detection unit, 40: Roadside information acquisition unit, 100: Action planning device, 120: Emergency stop request unit, 140: Stop position determination unit, 160: Action plan generation unit, 200: Vehicle control unit, 300: Information acquisition unit, 1000: Vehicle control system, 1001: Arithmetic processing circuit, 1002: Storage device, 1003: Input / output circuit, 1004: Communication circuit, 2000: Drive system device, OV: Vehicle, RU: Roadside unit, RUA: Area, CNF: Area, VE: Other vehicle, SN1,SN2,SN3,SN4,SN5,SN6,SN7:Detection range, SL,SL1,SL2,SL3,SL4:Stop line, SP,MRP:Stop position, STR Straight route, LTR: Left turn route, RTR: Right turn route, PD,PD1,PD2: Pedestrians, RW,RW1,RW2: Road width, LFW: Remaining route width, FSP: Area, ESP: Area, OL Other lanes.

Claims

1. Obstacle information around the moving object acquired by the obstacle information acquisition unit; Position information of the moving object acquired by a self-position acquisition unit; Road information around the moving object acquired by a road information acquisition unit, and an action planning device to which obstacle information around a roadside acquired by a roadside information acquisition unit is input, the action planning device including an emergency stop request unit that outputs an emergency stop request when an abnormality is detected in at least one of the obstacle information around the moving object acquired by the obstacle information acquisition unit and the obstacle information around the roadside acquired by the roadside information acquisition unit; a stop position determination unit that determines whether the moving object will stop within a first area that would hinder the progress of other traffic participants when the moving object stops, based on the road information, the position information of the moving object, and the emergency stop request output from the emergency stop request unit; and a behavior plan generation unit that generates a behavior plan for avoiding the moving body from stopping within the first area when the stop position determination unit determines that the moving body will stop within the first area where it will hinder the progress of other traffic participants.

2. The action plan generation unit a candidate route generating unit that generates at least one candidate route along which the moving object travels, based on the road information and the position information of the moving object; a route travel difficulty calculation unit that calculates a travel difficulty level for the moving body when traveling along each of the candidate routes generated by the candidate route generation unit, based on the obstacle information acquired by the obstacle information acquisition unit, the obstacle information around the roadside acquired by the roadside information acquisition unit, the position information of the moving body, and the abnormality information detected by the emergency stop request unit; and a behavior determination unit that selects a travel route with the lowest travel difficulty level for each of the candidate routes calculated by the route travel difficulty calculation unit, The behavior planning apparatus according to claim 1 , wherein a behavior plan is generated in which the moving object stops outside the first area.

3. The route travel difficulty calculation unit is If an obstacle exists beyond the first area, a second area is calculated as a stopping area required for the moving body to stop in front of the obstacle; The action planning device according to claim 2 , wherein the travel difficulty level is calculated to be high when at least a portion of the second area exists within the first area.

4. The candidate route generation unit A route along which the moving object is to travel to a target position is generated as one of candidate routes; The action determination unit is 4. The behavior planning device according to claim 2, wherein, when the travel difficulty levels are the same, a route along which the moving object is scheduled to move to a target position is preferentially selected.

5. An information storage unit that stores obstacle information around the roadside acquired by the roadside information acquisition unit, The action plan generation unit 4. The behavior planning device according to claim 1, wherein, when the stop position determination unit determines that the vehicle will stop in the first area where the stop will hinder the progress of other traffic participants, a behavior plan is generated to avoid the vehicle stopping in the first area by adding the obstacle information accumulated in the information accumulation unit.

6. An information storage unit that stores obstacle information around the roadside acquired by the roadside information acquisition unit, The route travel difficulty calculation unit is 4. The behavior planning device according to claim 2, further comprising: a step of calculating a degree of difficulty of travelling when the moving object travels along each of the candidate routes generated by the candidate route generating unit, by adding the obstacle information accumulated by the information accumulating unit.

7. The action plan generation unit 6. The behavior planning device according to claim 5, wherein, when the emergency stop request unit detects an abnormality in the obstacle information around the roadside acquired by the roadside information acquisition unit, the emergency stop request unit generates a behavior plan to avoid stopping within the first area by using the obstacle information accumulated in the information accumulation unit, without using the obstacle information around the roadside acquired by the roadside information acquisition unit.

8. The action planning device according to claim 1 , further comprising: the obstacle information acquisition unit, the self-position acquisition unit, the road information acquisition unit, and the roadside information acquisition unit.

9. A vehicle control system comprising: the behavior planning device according to claim 1; and a vehicle control unit that controls a moving object.

10. A behavior plan generating method executed by using a behavior planning device, comprising: acquiring obstacle information around the moving object; acquiring location information of the moving object; acquiring road information around the moving object; Obtaining obstacle information around a roadside; outputting an emergency stop request when an abnormality is detected in at least one of the acquired obstacle information around the moving object and the acquired obstacle information around the roadside; determining whether the moving body will stop in an area that would hinder the progress of other traffic participants when stopping the moving body, based on the acquired road information and position information of the moving body and the output emergency stop request; When it is determined that the moving body will stop within the area, generating an action plan for avoiding the moving body from stopping within the area.

11. In the step of generating an action plan, generating at least one candidate route along which the mobile object will travel based on the road information and the position information of the mobile object; calculating a travel difficulty level for the moving body when the moving body travels along each of the candidate routes based on obstacle information around the moving body, obstacle information around the roadside, position information of the moving body, and detected abnormality information; selecting a travel route with the lowest travel difficulty level from among the calculated travel difficulty levels; The method for generating a behavior plan according to claim 10, further comprising: generating a behavior plan for stopping the moving body outside the area based on a selected travel route.

12. After the step of acquiring obstacle information around the roadside, The method further includes a step of accumulating the obtained obstacle information around the roadside, In the step of generating an action plan, The method for generating a behavior plan according to claim 10 or 11, further comprising the step of generating a behavior plan for the moving object by adding the accumulated obstacle information in the vicinity of the roadside.

Images