Vehicle control device and vehicle control method

Abstract

A vehicle control device includes: a recognition unit that recognizes a surrounding situation based on a detection result of an external sensor; a travel trajectory generation unit that generates a travel trajectory from the current position to the target position on the basis of the recognition by the recognition unit; and a shiftable position determination unit that determines a shiftable position at which a shift from the automatic driving mode to the manual driving mode is possible during the execution of the driving task in the automatic driving mode. The travel trajectory generation unit generates a first travel trajectory, which is a travel trajectory from a current position to a target position, and a second travel trajectory, which is a travel trajectory from a point on the first travel trajectory to a shiftable position.

Description

Technical Field

[0001] The present invention relates to a vehicle control device and a vehicle control method. Background Art

[0002] In recent years, the development of autonomous driving technology in vehicles such as automobiles has been advancing. When performing autonomous driving, the roads on which vehicles travel do not always have an appropriate road width. For example, on a narrow road with a narrow travel lane width, or on a road where it is difficult for the host vehicle to pass by an oncoming vehicle, it is necessary to perform control different from normal driving in order to avoid to a place where passing is possible. Patent Document 1 discloses the following technology: When traveling on a road where there is no space to avoid, actions such as the departure, stop, and traveling direction of the host vehicle are planned to control the vehicle to return to the entrance of the narrow road. Prior Art Documents Patent Document

[0003] Patent Document 1: Japanese Patent Application Laid-Open No. 2018-151177 Summary of the Invention Technical Problem to be Solved by the Invention

[0004] In the previously proposed autonomous driving technology, when performing autonomous driving during passing on a narrow road in an urban area or the like, due to reasons such as a dead angle of a sensor after starting to drive for avoidance, it is sometimes determined that control is impossible. In this case, the control device determines that autonomous driving cannot be performed and gives up. When the control device gives up, the authority is transferred to the driver from the place where the control device gives up, but there are also cases where it becomes difficult for the driver to drive, and even manual driving is impossible. In addition, even if autonomous driving continues, there is a possibility of giving up when the vehicle cannot exit after entering the road to be avoided. As a result, the ride comfort and convenience of the vehicle deteriorate.

[0005] The present invention has been made in view of the above circumstances, and an object thereof is to provide a vehicle control device and a vehicle control method that can appropriately continue autonomous driving when passing on a narrow road or the like. Technical Means for Solving the Technical Problem

[0006] In order to solve the above problems, the structure described in the claims is adopted, for example. This application includes a plurality of methods for solving the above problems. If one example is cited, as a vehicle control device, a vehicle control device that controls a vehicle that transfers between an autonomous driving mode in which at least a part or all of the driving tasks can be implemented in a vehicle system and a manual driving mode in which a driver or a remote monitoring person implements the driving tasks is applied. The vehicle control device includes: an identification unit that identifies the surrounding conditions of the vehicle based on the detection results of external sensors equipped on the vehicle; a travel trajectory generation unit that generates a travel trajectory from the current position of the vehicle to a target position based on the surrounding conditions of the vehicle identified by the identification unit; and a transferable position determination unit that determines a transferable position at which the driving mode can be transferred from the autonomous driving mode to the manual driving mode during the execution of a driving task in the autonomous driving mode. Here, the travel trajectory generation unit separately generates a travel trajectory of the vehicle from the current position to the target position, i.e., a first travel trajectory, and a travel trajectory from a point on the first travel trajectory to the transferable position, i.e., a second travel trajectory, and controls the vehicle using the first travel trajectory or the second travel trajectory. Advantages of the Invention

[0007] According to the present invention, in a state where it is determined that automatic passing driving is impossible, an appropriate alternative trajectory can be planned and autonomous driving can be continued. Other technical problems, structures, and effects will be further clarified through the following description of the embodiments. Brief Description of the Drawings

[0008] Figure 1 It is a block diagram showing a structural example of a travel drive system and sensors of a vehicle equipped with the vehicle control device according to Embodiment 1 of the present invention. Figure 2 It is a block diagram showing a structural example of the vehicle control device according to Embodiment 1 of the present invention. Figure 3 It is a block diagram showing a structural example of a retention risk map generation unit of the vehicle control device according to Embodiment 1 of the present invention. Figure 4 It is a flowchart showing an example of a retention risk map generation process performed by the vehicle control device according to Embodiment 1 of the present invention. FIG. 5 is a diagram showing an example of passing driving of the vehicle control device according to Embodiment 1 of the present invention during driving on a narrow road ( Figure 5A : Example 1, Figure 5B : Example 2). Figure 6 It is a block diagram showing a structural example of a vehicle driving plan unit of the vehicle control device according to Embodiment 1 of the present invention. Figure 7 It is a block diagram showing a structural example of a passing trajectory planning unit of the vehicle control device according to Embodiment 1 of the present invention. Figure 8 It is a flowchart showing an example of a standby position and attitude generation process performed by the vehicle control device according to Embodiment 1 of the present invention. Figure 9 This is a diagram showing a control example of the parking position attitude obtained by the vehicle control device according to Example 1 of the embodiment of the present invention. Figure 10 This is a flowchart showing an example of the oncoming vehicle trajectory generation process performed by the vehicle control device according to Example 1 of the embodiment of the present invention. Figure 11 This is a diagram showing an example of the end position candidate obtained by the vehicle control device according to Example 1 of the embodiment of the present invention. Figure 12 This is a diagram showing an example of the oncoming vehicle trajectory obtained by the vehicle control device according to Example 1 of the embodiment of the present invention. Figure 13 is a diagram showing an example of the movement of the vehicle and the planned oncoming vehicle trajectory obtained by the vehicle control device according to Example 1 of the embodiment of the present invention ( Figure 13A : Example before entering the oncoming vehicle trajectory, Figure 13B : Example of the oncoming vehicle trajectory). Figure 14 is a diagram showing an example of the movement of the vehicle and the planned alternative trajectory obtained by the vehicle control device according to Example 1 of the embodiment of the present invention ( Figure 14A : Example of generating a coordinated action trajectory, Figure 14B : Example of being unable to follow the coordinated action trajectory). Figure 15 This is a flowchart showing an example of the driver replacement position generation process when giving up, which is performed by the vehicle control device according to Example 1 of the embodiment of the present invention. Figure 16 This is a flowchart showing an example of the alternative trajectory generation process when giving up, which is performed by the vehicle control device according to Example 1 of the embodiment of the present invention. Figure 17 This is a block diagram showing a structural example of the driving mode management unit of the vehicle control device according to Example 1 of the embodiment of the present invention. Figure 18 is a diagram showing an example of the movement of the vehicle and the planned alternative trajectory obtained by the vehicle control device according to Example 2 of the embodiment of the present invention ( Figure 18A : Example of the movement of the vehicle, Figure 18B : Example of the alternative trajectory). Figure 19 is a diagram showing an example of the driver replacement position when giving up, which is obtained by the vehicle control device according to Example 2 of the embodiment of the present invention ( Figure 19A : Example of the trajectory, Figure 19B : Example of the driver replacement position). Figure 20 is a diagram showing an example of the movement of the vehicle and the detection range of the vehicle-mounted sensors obtained by the vehicle control device according to Example 3 of the embodiment of the present invention ( Figure 20A : Example 1, Figure 20B : Example 2,Figure 20C : Diagram of Example 3). Figure 21 shows an example of the movement of the vehicle and the planned alternative trajectory obtained by the vehicle control device according to Embodiment Example 3 of the present invention ( Figure 21A : Example 1, Figure 21B : Diagram of Example 2). Figure 22 is a flowchart showing an example of the alternative trajectory generation process at the time of abandonment performed by the vehicle control device according to Embodiment Example 3 of the present invention. Detailed implementation mode

[0009] <Embodiment Example 1> Hereinafter, with reference to Figures 1 - 17 the vehicle control device and the vehicle control method according to Embodiment Example 1 of the present invention will be described.

[0010] [Structure of the vehicle] Figure 1 Shows the overall structure of the vehicle 100 equipped with the vehicle control device 1 according to Embodiment Example 1 of the present invention. The vehicle 100 performs autonomous driving under the control of the vehicle control device 1. As Figure 1 shown, the vehicle 100 has a left front wheel 101FL, a right front wheel 101FR, a left rear wheel 101RL, and a right rear wheel 101RR. The vehicle 100 is equipped with a front recognition sensor 2, a left side recognition sensor 3, a right side recognition sensor 4, and a rear recognition sensor 5 as sensors for recognizing the outside world, and can detect the relative distance and relative speed between the own vehicle and surrounding vehicles. Information (detection signals) from these sensors 2, 3, 4, and 5 is provided to the vehicle control device 1.

[0011] The vehicle control device 1 calculates command values for the steering control mechanism 10, the brake control mechanism 13, and the throttle control mechanism 20 for controlling the traveling direction of the vehicle 100 based on the information from the sensors 2, 3, 4, and 5. In addition, the vehicle 100 includes: a steering control device 8 that controls the steering control mechanism 10 based on the command value from the vehicle control device 1; and a brake control device 15 that controls the brake control mechanism 13 based on the command value from the vehicle control device 1 and adjusts the braking force distribution of each wheel. And the vehicle 100 includes: an acceleration control device 19 that controls the throttle control mechanism 20 based on the command value from the vehicle control device 1 and adjusts the torque output of the engine; and a display device 24 that displays the driving plan of the vehicle 100 and the action prediction of moving objects existing in the vicinity, etc.

[0012] In addition, the vehicle 100 is equipped with a communication device 23 for performing vehicle-to-road or vehicle-to-vehicle communication. In addition, Figure 1The illustrated sensor structure is an example, but not limited to Figure 1 the example of Figure 1 . Regarding the types of sensors, it can also be various sensors such as ultrasonic sensors, stereo cameras, infrared cameras, lasers, lidars (LiDAR), etc., or a combination of these sensors.

[0013] The vehicle control device 1 is configured as an arithmetic processing device having, for example, a CPU (Central Processing Unit) and a memory as described later in Figure 2 Figure 2 . Then, a program for performing vehicle driving control processing is stored in the vehicle control device 1, and a driving plan is generated by executing this program. The vehicle control device 1 calculates command values for controlling the actuators 10, 13, 20 for vehicle driving according to the generated driving plan. The control devices 8, 15, 19 of the respective actuators 10, 13, 20 receive the command values of the vehicle control device 1 through communication, and control the respective actuators based on these command values.

[0014] When explaining the structure for actuating the Figure 1 actuator shown in Figure 1 , when the driver is driving, the stepping force generated by the driver stepping on the brake pedal 12 is multiplied by a brake booster (not shown), and a corresponding oil pressure is generated by a master cylinder (not shown). The generated oil pressure is supplied to 16FL, 16FR, 16RL, 16RR disposed on each wheel 101FL to 101RR via the brake control mechanism 13.

[0015] The wheel cylinders 16FL to 16RR are composed of a cylinder block, a piston, brake pads, etc. The piston is pushed by the working fluid supplied from the master cylinder 9, and the brake pads connected to the piston are pressed against the disc rotor. In addition, the disc rotor rotates together with the wheels constituting each wheel 101FL to 101RR. Therefore, the braking torque acting on the disc rotor becomes the braking force acting between the wheel and the road surface. With the above structure, braking force can be generated on each wheel according to the driver's brake pedal operation.

[0016] Similar to the vehicle control device 1, the brake control device 15 is configured as an arithmetic processing device having, for example, a CPU and a memory. Sensor signals from a composite sensor 14 capable of detecting longitudinal acceleration, lateral acceleration, and yaw rate, sensor signals from wheel speed sensors 11F to 11RR provided on each wheel, braking force commands from the brake control device 15, and sensor signals from a steering wheel angle detection device 21 via a later-described steering control device 8 are input to the brake control device 15. In addition, the output of the brake control device 15 is connected to a brake control mechanism 13 having a pump and a control valve, and arbitrary braking force can be generated on each wheel independently of the driver's brake pedal operation.

[0017] The braking control device 15 estimates the rotation, drift, and wheel lock of the vehicle based on the input of sensor signals, and generates braking force on the corresponding wheels to suppress these situations, thereby improving the handling stability of the driver. In addition, the vehicle control device 1 sends a braking command to the braking control device 15, so that the vehicle 100 generates arbitrary braking force to automatically perform braking in the autonomous driving without the driver's operation. However, the mechanism for braking is not limited to this structure, and other actuators such as brake by wire can also be used.

[0018] Next, the structure for performing the steering operation will be described. In the state where the driver is driving the vehicle, the steering torque detection device 7 detects the steering torque input by the driver via the steering wheel 6, and the steering wheel angle detection device 21 detects the steering wheel angle. Based on this information, the steering control device 8 controls the motor to generate auxiliary torque. In addition, similarly to the vehicle control device 1, the steering control device 8 is also configured as an arithmetic processing device including a CPU and a memory, for example. The steering control mechanism 10 is movable by the resultant force of the driver's steering torque and the auxiliary torque of the motor, so that the front wheels are steered. On the other hand, it is configured that according to the steering angle of the front wheels, the reaction force from the road surface is transmitted to the steering control mechanism and is transmitted to the driver as the road surface reaction force.

[0019] The steering control device 8 can generate torque by the motor independently of the driver's steering operation and control the steering control mechanism 10. Therefore, the vehicle control device 1 can control the front wheels to an arbitrary steering angle by communicating a steering force command to the steering control device 8. Therefore, it plays a role in automatically steering in the autonomous driving without the driver's operation. However, the present embodiment is not limited to Figure 1 the shown steering control device 8, and it can also be configured to use other actuators such as steer by wire to perform the steering operation.

[0020] Next, the structure of the accelerator will be described. The depression amount of the accelerator pedal 17 by the driver is detected by the stroke sensor 18 and input to the acceleration control device 19. In addition, similarly to the vehicle control device 1, the acceleration control device 19 is also configured as an arithmetic processing device including a CPU and a memory, for example. The acceleration control device 19 adjusts the throttle opening according to the depression amount of the accelerator pedal 17 and controls the engine. Thus, the acceleration control device 19 can accelerate the vehicle according to the driver's accelerator pedal operation. In addition, the acceleration control device 19 can control the throttle opening independently of the driver's acceleration operation. Therefore, the vehicle control device 1 can cause the vehicle to generate an arbitrary acceleration by communicating an acceleration command to the acceleration control device 19. Therefore, it plays a role in automatically accelerating during autonomous driving in which the driver does not operate.

[0021] [Structure of Vehicle Control Device] Next, regarding the structure of the vehicle control device 1 that performs autonomous driving control installed in the vehicle control system of this embodiment, it will be described using Figure 2 as an example. As Figure 2 shown on the outside of the block diagram, the vehicle control device 1 is constituted by, for example, a computer as an information processing device. That is, the computer of the vehicle control device 1 includes a CPU 211 as a processor, a memory 212, an input / output unit 213, and an interface 214.

[0022] As the memory 212, in addition to memories such as ROM (Read Only Memory) and RAM (Random Access Memory), storage devices such as HDD (Hard Disk Drive) and SSD (Solid State Drive) are also used. In the memory 212, information such as programs for operating as the vehicle control device 1, driving conditions, and histories is stored. In addition, by controlling the CPU 211 to read and execute the program, the processing function units 201 to 205 described later are configured in the memory 212.

[0023] The input / output unit 213 performs Figure 1 input processing of information from sensors 2 to 5 as shown, and output processing of command values for each actuator 10, 13, 20. The interface 214 performs information transmission processing with other information processing devices in the vehicle 100 such as the brake control device 15, and performs transmission and reception processing with the outside via Figure 1 the communication device 23 as shown.

[0024] When describing the processing functional units configured in the vehicle control device 1, an automatic driving planning unit 201, an automatic parking planning unit 202, a vehicle motion control unit 203, an actuator control unit 204, and a dwelling risk map generation unit 205 are configured in the vehicle control device 1. These processing units 201 to 205 can communicate with each other via the vehicle network 206. In addition to wired connections, the vehicle network 206 sometimes uses wireless connections.

[0025] The automatic driving planning unit 201 plans the operation of the host vehicle for automatically driving the vehicle 100 to the destination. The automatic parking planning unit 202 plans the operation of the host vehicle for automatically parking the vehicle 100 in a parking space or the like into a parking frame. The vehicle motion control unit 203 generates command values for controlling the motion of the vehicle 100 during automatic driving. The actuator control unit 204 controls various actuators such as an engine, a brake, and a steering gear. The dwelling risk map generation unit 205 generates a dwelling risk map of the vehicles existing around the host vehicle including the target vehicle and the like. In addition, the actuator control unit 204 is sometimes installed on hardware different from the vehicle control device 1, such as an engine control controller and a brake control controller.

[0026] [Structure of dwelling risk map generation unit] Figure 3 The structure of the dwelling risk map generation unit 205 of the vehicle control device 1 is shown. Information from the radar 301, the stereo camera 302, the vehicle sensor 303, and the lidar 304 is provided to the dwelling risk map generation unit 205. That is, as a sensor for recognizing the outside world, the radar 301 can emit radio waves to an object and measure its reflected wave, thereby measuring the distance and direction to the object. Information on the distance and direction to the object obtained by the radar 301 is provided to the dwelling risk map generation unit 205. In addition, the stereo camera 302 can obtain information on its depth direction by simultaneously photographing an object from multiple different directions. The information on the depth direction obtained by the stereo camera 302 is provided to the dwelling risk map generation unit 205.

[0027] In addition, the vehicle sensor 303 is a sensor that obtains the following information: information on the speed of the vehicle and the rotational speed of the tires; information obtained by calculating the average position of an autonomous driving vehicle using GNSS (Global Navigation Satellite System); destination information input by a person riding in the autonomous driving vehicle through an interface of a navigation system; and destination information specified by a remote operator or the like using wireless communication such as a telephone line. The information obtained by these vehicle sensors 303 is provided to the stay risk map generation unit 205. The lidar 304 measures the scattered light of the radar irradiation for pulsed light emission and detects the distance to an object located at a long distance. The information on the distance to the object located at a long distance obtained by this lidar 304 is provided to the stay risk map generation unit 205.

[0028] The stay risk map generation unit 205 includes a sensor information processing unit 305, a map information processing unit 306, a three-dimensional object motion prediction unit 307, a storage unit 308, a stay risk map calculation unit 309, and a self-position estimation processing unit 310. The detection information obtained by the external sensors including the radar 301, the stereo camera 302, the vehicle sensor 303, and the lidar 304 is input to the sensor information processing unit 305. Thereby, the sensor information processing unit 305 obtains the object information of the moving objects existing around the own vehicle from the detection information of the external sensors.

[0029] As specific object information, attribute information such as pedestrians, bicycles, and vehicles, as well as their current positions and current velocity vectors are extracted. Here, as moving objects, parked vehicles and the like that may move in the future even if the speed obtained at the current moment is zero are included. The sensor information processing unit 305 functions as an identification unit, and the sensor information processing unit 305 performs an identification process for identifying the situation around the vehicle 100 based on the information of the detection results of the external sensors provided in the vehicle.

[0030] In the storage unit 308, road information and signal light information from the starting point of autonomous driving of the own vehicle to the target point and its surroundings, path information from the current position to the target point, a traffic rule database for the traveled section, etc. are stored. In addition, in the storage unit 308, a point group database used in the self-position estimation processing unit 310 is also stored. The map information processing unit 306 obtains the lane center line information and signal light information of the roads required for autonomous driving from the information stored in the storage unit 308. Then, the map information processing unit 306 organizes the lighting information of the signal lights that the autonomous driving vehicle is scheduled to pass through, etc., and sets it as information in a form that can utilize these lane center line information and signal light information. Furthermore, the own - position estimation processing unit 310 estimates the location where the host vehicle exists based on the surrounding information obtained from the vehicle sensor 303, etc., the point - group data, the steering angle of the vehicle, the vehicle speed, and the information obtained from GNSS.

[0031] Then, the output of the sensor information processing unit 305, the output of the map information processing unit 306, and the output of the own - position estimation processing unit 310 are provided to the moving - object motion prediction unit 307. Based on these input information, the moving - object motion prediction unit 307 calculates the future position and speed information (object prediction information) of each moving object. To predict the movement of each moving object, the moving - object motion prediction unit 307 first predicts the position R(X(T), Y(T)) of each object at the future time T based on the object information. As a specific prediction method, when the current position of the moving object is set as Rn0(Xn(0), Yn(0)) and the current speed is set as Vn(Vxn, Vyn), the moving - object motion prediction unit 307 performs prediction calculations based on the following linear prediction formula, i.e., "Mathematical Formula 1" for example.

[0032]

Mathematical Formula 1

[0033] The calculation of "Mathematical Formula 1" is performed under the assumption that each object moves in a uniform linear motion maintaining the current speed within the future time. Thus, more objects can be predicted in a short time. In addition, as another calculation method, the position and speed information of other vehicles obtained from sensors, etc., or the image information obtained from a camera can be input into a neural network model with learning completed, so as to obtain the prediction result of the position or speed information of other vehicles at the future time. According to this calculation, the reliability of the prediction result can also be calculated.

[0034] The stay - risk map calculation unit 309 generates a stay - risk map for surrounding vehicles such as oncoming vehicles through calculation processing. When there is a scenario where a collision or stalemate will occur if no coordinated action is taken with objects such as vehicles, pedestrians, and bicycles existing in the surrounding area, the stay - risk map calculation unit 309 generates a stay - risk map using the predicted trajectory calculated based on the target actions of the objects of the coordinated - action objects, and outputs the generated stay - risk map.

[0035] [Stay - risk map generation process] Figure 4 It is a flowchart showing an example of the process in which the stay - risk map generation unit 205 generates a stay - risk map. First, the dwelling risk map generation unit 205 acquires various types of information including surrounding information (step S11). Next, the dwelling risk map generation unit 205 predicts the current position and speed information of oncoming vehicles etc. that need to perform passing maneuvers in cooperation with the host vehicle based on the acquired surrounding information, and predicts the target actions of the target vehicle based on the road environment information during the driving of the host vehicle and the target vehicle (step S12). Additionally, as the road environment information, road shape, object information existing in the surroundings, surrounding map information, congestion information, etc. can be considered.

[0036] Then, the dwelling risk map generation unit 205 determines whether the target vehicle is a moving object that needs to cooperate with the host vehicle (step S13). When it is determined in step S13 that cooperation is required (step S13 is YES), the dwelling risk map generation unit 205 generates and outputs a cooperation action plan and a dwelling risk map based on the target actions of the target vehicle that requires cooperation (step S14). Furthermore, when it is determined in step S13 that cooperation is not required (step S13 is NO), the dwelling risk map generation unit 205 does not generate a dwelling risk map and ends the process.

[0037] Figure 5A and Figure 5B Examples (Example 1 and Example 2) showing the generation results of the cooperation action plan and the dwelling risk map are presented. Figure 5A Example (Example 1) of Figure 5B Example (Example 1) of both show a situation where the host vehicle M1 and the oncoming vehicle M2 of the target that requires cooperation perform a passing maneuver on a narrow road. Figure 5A Example of is a case where the road boundary 741 on the left side of the host vehicle M1 is non-linear and has an avoidance road, Figure 5B Example of is a case where the road boundary 741 on the left side of the host vehicle M1 is linear and the avoidance road is on the opposite side (right side when observed from the host vehicle M1).

[0038] Thus, in Figure 5A or Figure 5B In the situation shown, the dwelling risk map generation unit 205 predicts the arrival target location 742 of the oncoming vehicle M2 at a future time (a specified time until the cooperation action is completed) based on various types of information. Then, the dwelling risk map generation unit 205 predicts the path for the oncoming vehicle to reach the predicted location and calculates the area when the oncoming vehicle passes through this path. The results calculated by the dwelling risk map generation unit 205 become the dwelling risk map 701 in Figure 5A Example of, Figure 5B Example of.

[0039] In addition, based on the arrival target location 742 and the residence risk map 701, the residence risk map generation unit 205 calculates the vehicle avoidance position 731 for the host vehicle M1 to avoid. In order to jointly complete the actions of the host vehicle M1 and the oncoming vehicle M2, the residence risk map generation unit 205 explores the sequence of executing each avoidance action serially or in parallel, and sets the sequence that can achieve it as the cooperative action plan. As the cooperative action plan, first, the residence risk map generation unit 205 formulates a plan for the host vehicle M1 to enter the vehicle avoidance position 731, and then formulates a plan for the oncoming vehicle M2 to pass by the side of the host vehicle M1. Finally, a plan for the host vehicle M1 to leave the avoidance road is formulated. In the case where a sequence to complete such actions cannot be found, the arrival target location, the residence risk map, and the vehicle avoidance position are reexamined, and the sequence is recursively searched.

[0040] [Structure of the Autopilot Planning Unit] Figure 6 An example of the structure of the autopilot planning unit 201 is shown. The autopilot planning unit 201 includes an autopilot plan generation unit 221 that calculates the target trajectory, a driving mode management unit 222, and a trajectory planning unit 230.

[0041] The residence risk map, the cooperative action plan, the lane information, the map information, the environment information, the path information, the UI information, etc. are provided to the autopilot plan generation unit 221. Based on the path information and the environment information, etc., the autopilot plan generation unit 221 calculates the weights of the target action candidates that the host vehicle can take. The weights of the target action candidates are the weights of the LK candidates for maintaining the current lane, the LC candidates for changing lanes from the current lane to an adjacent lane, the OA candidates for avoiding obstacles in front, the CO candidates for cooperative actions with other vehicles, etc. for the actions that the host vehicle can take. For example, when driving on a straight road, considering that there are no vehicles or objects to avoid in front and no lane change to an adjacent lane is required from the path information, in this situation, the weight of LK = 100, the weight of LC = 0, the weight of OA = 0, the weight of CO = 0, etc.

[0042] The trajectory planning unit 230 includes a lane keeping trajectory generation unit (LK) 231, a lane change trajectory generation unit (LC) 232, an obstacle avoidance trajectory generation unit (OA) 233, a cooperative action trajectory generation unit (CO) 234, and a trajectory mediation unit 235. The trajectory planning unit 230 functions as a driving trajectory generation unit, and performs a driving trajectory generation process for generating a driving trajectory from the current position of the vehicle to the target position. The lane keeping trajectory generation unit 231 generates a trajectory for keeping the vehicle at the center of the lane in which it is currently traveling. The lane change trajectory generation unit 232 generates a trajectory for changing lanes to an adjacent lane to the lane in which the vehicle is currently traveling.

[0043] The obstacle avoidance trajectory generation unit 233 generates a trajectory for avoiding an object that obstructs travel and exists in the lane in which the vehicle is currently traveling. The cooperative action trajectory generation unit 234 generates a trajectory for performing a cooperative action with surrounding objects. For the lane keeping trajectory, the lane change trajectory, the obstacle avoidance trajectory, and the cooperative action trajectory, the trajectory mediation unit 235 evaluates each trajectory based on the safety level with surrounding objects and the target action candidate weight, selects the trajectory with the best evaluation, and generates a target trajectory.

[0044] The driving mode management unit 222 calculates the previous selection information for calculating the target action candidate weight at the next sampling time based on the target driving mode selected by the trajectory mediation unit 235, the trajectory evaluation values based on each action candidate, and the manual driving request in the cooperative action trajectory generation unit 234. For example, in the case where LK (keep the current lane: lane keeping) is selected with evaluation values of LK = 60, LC = 40, OA = 0, and C = 0, the previous selected current driving information is generated so that the possibility of selecting LK at the next sampling time is increased for the continuity of the action.

[0045] [Structure of the Cooperative Action Trajectory Generation Unit] Figure 7 Shows the structure of the cooperative action trajectory generation unit 234. The cooperative action trajectory generation unit 234 includes a cooperative action state management unit 241, a standby position and attitude generation unit 242, a cooperative action trajectory calculation unit 243, a driving replacement position generation unit 244 at the time of abandonment, a standby trajectory generation unit 245, and a cooperative action trajectory selection unit 246. The stay risk map, the cooperative action plan information, the environment information, the lane information, and the map information are provided to the cooperative action trajectory generation unit 234. The cooperative action state management unit 241 determines whether it is possible to follow the cooperative action trajectory, and determines that it is in a stalemate mode when it is impossible to follow. The standby position and attitude generation unit 242 generates candidate target standby positions and attitudes in the cooperative action plan. The cooperative action trajectory calculation unit 243 calculates the driving trajectory (first driving trajectory) in the cooperative action mode.

[0046] The driving replacement position generation unit 244 at the time of abandonment generates the driving replacement position on the driving trajectory. The driving replacement position is, for example, a position where a mode transfer from the autonomous driving mode to the manual driving mode can be performed, and the driving replacement position generation unit 244 functions as a transferable position determination unit. The alternative trajectory generation unit 245 generates an alternative trajectory (second driving trajectory) to the driving replacement position when it is impossible to drive on the driving trajectory in the cooperative action mode. The cooperative action trajectory selection unit 246 selects an appropriate trajectory from the cooperative action trajectory calculated by the cooperative action trajectory calculation unit 243 and the alternative trajectory generated by the alternative trajectory generation unit 245, and controls the driving of the vehicle 100.

[0047] Figure 8 It is a flowchart showing an example of the standby position attitude generation process performed by the standby position attitude generation unit 242 in the cooperative action trajectory generation unit 234. First, the standby position attitude generation unit 242 acquires various information such as a stay risk map, cooperative action plan information, environmental information, lane information, and map information (step S21). Next, the standby position attitude generation unit 242 generates a target standby position attitude candidate in the current cooperative action step in the cooperative action plan (step S22).

[0048] Figure 9 It shows an example of the generation process of the target standby position attitude candidate in step S22. Figure 9 In the example, there is an oncoming vehicle M2 to perform a cooperative action in front of the own vehicle M1. Also, there is an avoidance space on the left front side of the own vehicle M1. Figure 9 It shows the road boundary 741. Figure 9 In the example, the stay risk map 701 generated by the stay risk map generation unit 205 exists in front of the own vehicle M1. Therefore, in order to achieve the cooperative action here, that is, passing by driving, the own vehicle M1 needs to move to a place where the stay risk map 701 does not exist and standby. Therefore, the stay risk map generation unit 205 generates avoidance attitude position candidates (N1, θ1) to (N5, θ5) that do not have the stay risk map 701 and are the drivable area (inside the road boundary 741) around. N represents the planar position coordinate, and θ represents the vehicle's heading angle. The number of avoidance attitude position candidates generated at this time is arbitrary, but as the generated place, a place close to the stay risk map 701 within the range considering the size of the own vehicle M1 is selected.

[0049] [System action trajectory generation process] Figure 10This is a flowchart showing the process of the cooperative action trajectory calculation unit 243 generating a cooperative action trajectory. First, the cooperative action trajectory calculation unit 243 acquires various information such as a residence risk map, cooperative action plan information, environment information, lane information, and map information (step S31). Then, the cooperative action trajectory calculation unit 243 generates an end position candidate, which is the arrival position where the driving mode is changed from the cooperative action mode to another mode after passing the avoidance position (step S32). After that, the cooperative action trajectory calculation unit 243 generates a trajectory plan candidate that passes through the avoidance position candidate and the end position candidate (step S33).

[0050] In addition, the cooperative action trajectory calculation unit 243 evaluates the generated trajectory candidates and selects the best trajectory (step S34). Finally, the cooperative action trajectory calculation unit 243 stores the cooperative action trajectory information selected in step S34 in the storage unit 308 (step S35).

[0051] Here, the process of evaluating the trajectory candidates generated in step S34 and selecting the best trajectory will be described below. In order to select the best trajectory from the generated trajectory candidates, as an evaluation operation method for each trajectory candidate, for example, an equation of "Mathematical Formula 2" is considered. As described above, trajectory candidates are generated for each avoidable attitude position candidate and end position candidate. Therefore, it is necessary to select the best candidate from these candidates. Therefore, for each trajectory candidate, the following evaluation function is used to set the comprehensive evaluation value of each trajectory candidate.

[0052] [Mathematical Formula 2]

[0053] In the formula of "Mathematical Formula 2", for the individual evaluation values of the five items, weight coefficients for each item are set, and the sum of the products of the weight coefficients and the individual evaluation values is calculated. As the five individual evaluation values, they are set as the evaluation of safety, convenience, ride comfort, sense of incongruity, and sense of oppression on oncoming vehicles. Regarding the evaluation of safety, it is evaluated whether sufficient distance is ensured to avoid contact or approach with surrounding objects when driving on the passing path. For this purpose, the cooperative action trajectory calculation unit 243 uses the distance from surrounding objects and the potential risk value to calculate the evaluation amount of safety. For example, the cooperative action trajectory calculation unit 243 considers methods such as setting the reciprocal of the minimum distance between the road and the path candidate or the value of the collision risk map as the evaluation value.

[0054] Regarding the evaluation of convenience, when there is an oncoming vehicle, if the coordinated action, here the action for meeting the vehicle, takes a long time, the oncoming vehicle will wait longer, and the other vehicles and the passengers of the vehicle may feel distrustful. Therefore, in order to make the speed of the vehicle constant regardless of the length of the path, the length of the avoidance path of the vehicle is used to calculate the convenience evaluation amount when evaluating the convenience. For example, the coordinated action trajectory calculation unit 243 may consider a method of using the length of the path on the warehousing side as the evaluation value.

[0055] As an evaluation of ride comfort, the evaluation is based on whether the acceleration or jerk generated by the vehicle when the vehicle follows the avoidance path becomes larger. For example, it can be considered that the ride comfort of a path with a longer lateral acceleration and a longer lateral acceleration generation time is poor. In addition, when the turning radius of the path is small, the larger the longer the turning angle, the larger the evaluation value of the ride comfort becomes, which can be regarded as poor ride comfort. Therefore, the collaborative action trajectory calculation unit 243 considers setting the value of (the inverse of the turning radius) × (the total turning angle) as the evaluation index.

[0056] Regarding the evaluation of the sense of discomfort, it is considered that passengers in the host vehicle will feel a sense of discomfort when the vehicle is on the road and needs to reverse, and the evaluation value is calculated using the number of times the vehicle reverses. As an evaluation of the sense of pressure on the oncoming vehicle, a path that is too close to the oncoming vehicle may cause passengers of the own vehicle and other vehicles to feel distrust or fear. Therefore, the collaborative action trajectory calculation unit 243 calculates the evaluation value based on the distance between the candidate path of the oncoming vehicle. For example, the collaborative action trajectory calculation unit 243 considers a method of using the reciprocal of the minimum distance between the center position coordinates of other vehicles and the path candidate as an evaluation index.

[0057] [End Position Candidate Generation Process] Figure 11 A diagram for explaining an example of the process of generating end position candidates in step S32. The end position here is a location where it is expected to shift from the cooperative action mode to another mode (such as the lane keeping mode) later. Figure 11 In the example of FIG. 1 , when the possible oncoming position is set by the road boundary 741, the oncoming mode start point and each avoidance posture position candidate are line-symmetrical on the center line 751 of the driving lane of the host vehicle M1 stored in the map information ( Figure 11 The position where the length of L is equal to that of L is set as the end position candidate.

[0058] [Processing of generating candidate collaborative action trajectories] Figure 12A diagram illustrating an example of the generation process of a cooperative action trajectory candidate in step S33. To generate a cooperative action trajectory, a cooperative action target path is generated, and then a cooperative action target speed for traveling on the generated path is generated. As a generation process of the cooperative action target path, for example, using a stay risk map and candidate positions of avoidable postures, arcs, straight lines, and spiral curves are generated for a target avoidance position and candidate end positions. Figure 12 In this case, regarding the path of the host vehicle M1 from the current position to the avoidance position, the road boundary 741 is linear, and the section up to the start of the avoidance position (section ds) is set as a straight line substantially parallel to the center line 751. Then, at the deformed part of the road boundary 741 (section dc), a curved path using the Figure 12 two arc curves shown is set as the path to the avoidance position (N3, θ3). After passing the oncoming vehicle M2, the cooperative action trajectory calculation unit 243 is set to return to the path on the center line 751.

[0059] Figure 13A and Figure 13B show examples of representative scenarios of cooperative action driving and cooperative action trajectories when there is an oncoming vehicle M2 in front of the host vehicle M1 ( Figure 13A : before entering the passing trajectory, Figure 13B : an example of the passing trajectory). Here, as Figure 13A shown, the host vehicle M1 and the oncoming vehicle M2 as a cooperative action object drive sandwiching the passing point. There is an avoidance area on the left side of the host vehicle M1, and there is an obstacle 706 in the dead angle area 700 of the sensors of the host vehicle M1. In addition, a stay risk map 701 calculated based on the surrounding environment and the configuration of the oncoming vehicle M2 is obtained. In this situation, as Figure 13B shown, based on the stay risk map 701, the cooperative action trajectory calculation unit 243 calculates the avoidance position 702 of the host vehicle M1, the trajectory 704 from the current position to the avoidance position 702, and the trajectory 705 from the avoidance position 702 to the end position 703.

[0060] [Case of being unable to follow the cooperative action trajectory] Figure 14A and Figure 14B show examples when it is impossible to follow the planned cooperative action trajectory due to an oncoming vehicle or the surrounding environment while traveling on the cooperative action trajectory. Here, as Figure 14AAs shown, the vehicle M1 realizes the coordinated action by following the coordinated action trajectories 704 and 705. During the course of the action, the vehicle M1 is in the following state: when planning the coordinated action trajectory, an obstacle 706 in the blind spot area 700 is detected. In addition, during the course of the action, the oncoming vehicle M2, which is the target of the coordinated action, is approaching the vehicle M1. Therefore, it is impossible to generate a path for the vehicle M1 to avoid the obstacle 706, and it is difficult for the other vehicle M2 to retreat due to the narrow road. Therefore, it becomes difficult for the two vehicles to continue the coordinated action.

[0061] In this Figure 14A In the situation shown, the vehicle control device 1 of the vehicle M1 of the present embodiment cannot perform a coordinated action with the oncoming vehicle M2 as a coordinated action target, and the coordinated action driving mode is terminated. After the vehicle becomes safe, the passenger or remote operator is notified that the coordinated action driving mode cannot be continued. This notification can be made at the time when the coordinated action driving mode is terminated.

[0062] In addition, while the vehicle control device 1 of the host vehicle M1 generates a cooperative action trajectory and avoids the position to be avoided, when the oncoming vehicle M2 of the cooperative action object approaches and the speed is zero considering the risk of collision, and the following vehicle ( Figure 14A Figure 14B (not shown) is approaching, the vehicle M1 becomes stuck and cannot move forward or backward. Figure 14B As shown, the vehicle M1 can sometimes retreat to the retreat position 711, but cannot retreat in this way when a following vehicle (not shown) is approaching.

[0063] When the stalemate continues for a certain period of time, it is determined that it is difficult to break the state with the vehicle M1 alone. Therefore, the vehicle control device 1 notifies the passenger or the remote operator at a remote location through a display device mounted on the vehicle, etc. that the vehicle M1 is in a stalemate, and urges the passenger or the remote operator to drive. However, in the case where the stalemate is notified and a manual driving request is issued, assuming a state of being close to the vehicle of the collaborative action object or surrounding objects, it is foreseeable that the passenger or the remote operator will feel burdened by performing manual driving from this state, which requires driving to avoid collision with surrounding objects and resolve the stalemate. Therefore, it is preferred that the passenger or the remote operator switch to manual driving operation after moving to a place where the stalemate can be resolved in a safer and more secure state. The vehicle control device 1 of this embodiment is Figure 7 The driving replacement position generating unit 244 and the backup trajectory generating unit 245 shown are used to generate appropriate backup trajectories to the position where the passenger replaces the driver in the event of a stalemate.

[0064] [Processing for Generating Driving Replacement Location at the Time of Abandonment] Figure 15 It is a flowchart showing the processing of the driving replacement location generation unit 244 for generating a driving replacement location at the time of abandonment. First, the driving replacement location generation unit 244 acquires various information such as a stay risk map, a cooperative action plan information, environmental information, lane information, and map information (step S41). Then, the driving replacement location generation unit 244 generates a driving replacement location at the time of abandonment (S42). In this embodiment example, the driving replacement location at the time of abandonment is set to the location where the cooperative action mode starts from the lane keeping mode, etc., or when the passenger starts the cooperative action mode, etc.

[0065] Thus, if the vehicle can be automatically driven to the location where the cooperative action mode starts and the passenger or the remote operator at a remote location is urged to perform driving replacement, it is considered that the driving burden on the driving replacer can be reduced. In addition, when the passenger, etc. manually drives to the location where the cooperative action starts, it returns to the initial state of oncoming vehicle driving assistance, so the driver's burden can be reduced.

[0066] [Standby Trajectory Generation Processing] Figure 16 It is a flowchart showing the processing of the standby trajectory generation unit 245 for generating a standby trajectory. First, the standby trajectory generation unit 245 acquires various information such as a stay risk map, a cooperative action plan information, environmental information, lane information, and map information (step S51). Then, the standby trajectory generation unit 245 generates standby trajectory candidates to the driving replacement location at the time of abandonment (step S52). The driving replacement location here is the location generated by the driving replacement location generation unit 244. The path to the driving replacement location is Figure 12 as described in, and is preferably generated by combining forward or backward, etc. to generate a plurality of them.

[0067] Next, when there are a plurality of candidates generated in step S52, the standby trajectory generation unit 245 evaluates the plurality of trajectory candidates and selects the best trajectory from the results (step S53). As the processing for evaluating the plurality of trajectory candidates, for example, the method described in step S34 of the flowchart of Figure 10 can be applied. After that, the standby trajectory generation unit 245 stores the selected standby trajectory in the storage unit 308 (step S54).

[0068] [Structure of Driving Mode Management Unit] Figure 17 Shows a structural example of the driving mode management unit 222 ( Figure 6 ). The driving mode management unit 222, as Figure 17 shown, manages three states: the manual driving mode 251, the system shutdown mode 252 of the automatic driving system, and the automatic driving mode 253. When the power switch of the vehicle 100 is turned on or the like, the driving mode management unit 222 transfers from the system shutdown mode 252 to the manual driving mode 251 or the like. In addition, when the driver or the like turns on the automatic driving switch or the like, the driving mode management unit 222 transfers from the manual driving mode 251 to the automatic driving mode 253.

[0069] In the automatic driving mode 253, as described above, there are lane keeping (LK) and lane change (LC), etc. In addition, as a state between them, there is a transition intermediate state. When a stalemate occurs in the state of the cooperative action mode (CO) and it is determined that the automatic driving system needs to give up, a standby trajectory is generated. Then, after automatically driving to the driving replacement position, the driving mode management unit 222 issues a manual driving request through the cooperative action trajectory selection unit 246 ( Figure 7 ) and transfers to the manual driving mode 251.

[0070] [Effects of Embodiment Example 1] As described above, according to the vehicle control device 1 of the present embodiment example, when it is necessary to meet an oncoming vehicle during automatic driving, the following processing is performed: generating a cooperative action trajectory based on the oncoming vehicle and the surrounding conditions, and at the same time generating a standby trajectory. Thus, the present vehicle can appropriately meet the oncoming vehicle without colliding with it. In addition, when generating the standby trajectory, based on generating multiple candidates, an appropriate trajectory is selected according to the comprehensive evaluation value of each trajectory candidate, so that the cooperative action can be executed without affecting the riding comfort and without giving passengers a sense of insecurity.

[0071] Here, in the case of the present embodiment example, the start position of the automatic driving mode based on the cooperative action is set as the mode transfer possible position, that is, the driving replacement position, so that it is possible to simply and appropriately return to the driving replacement position.

[0072] <Embodiment Example 2> Next, with reference to FIGS. 18 to 19, the vehicle control device and the vehicle control method according to Embodiment Example 2 of the present invention will be described. In Embodiment Example 2, the structure of the vehicle control device 1 and the like provided in the vehicle 100 is the same as the structure described in Embodiment Example 1. The difference from Embodiment Example 1 is that the standby trajectory generation unit 245 generates a trajectory by setting the driving replacement position at the time of giving up to a location other than the meeting start position.

[0073] [Example of Setting a Position Other than the Meeting Start Position as the Driving Replacement Position] Figure 18A andFigure 18B In the example of Figure 18A , the following state exists: During the process of the host vehicle M1 calculating the trajectory 704 from the current position to the position 702 to be avoided and the trajectory 705 from the position 702 to be avoided to the end position 703 and moving, an obstacle 706 is detected. Here, the following state exists: Behind the host vehicle M1, a following vehicle M3 is approaching, and it is impossible to return to the passing start position.

[0074] In this case, the alternative trajectory generation unit 245 performs the following processing as Figure 18B shown: Search for a position 722 where driving replacement can be safely performed at a location other than the passing start position, and generate an alternative trajectory 721 to this location.

[0075] [Specific calculation example of driving replacement position] Figure 19A and Figure 19B show an example of the process in which the alternative trajectory generation unit 245 searches for and sets the driving replacement position 722. First, as Figure 19A shown, the following state exists: The host vehicle M1 detects an obstacle 706 during the process of following the cooperative action trajectory, and generates an alternative trajectory from the current position of the host vehicle M1. Or, it is in a state where an alternative trajectory is generated before the obstacle 706 is detected.

[0076] In this state, the following vehicle M3 is approaching from behind the host vehicle M1, and it is impossible to return to the start point of the cooperative action mode. Therefore, in this embodiment example, on the narrow road where the host vehicle M1 travels, the alternative trajectory generation unit 245 searches for the driving replacement position based on the lane center line 751 set in the map information carried by the host vehicle M1 and the dead angle information of the vehicle sensor. For example, the alternative trajectory generation unit 245 sets the driving replacement position 722 at the center position 751 in the longitudinal direction inside the path 753 to be avoided on the lane center line 751.

[0077] In addition, a position where the driving risk of the vehicle M1 determined at least according to the dead angle area of the external sensor of the host vehicle M1 or the presence or absence of an obstacle is below a threshold value can be considered as the driving replacement position 722 (mode transfer possible position). As a calculation method of the driving risk, for example, as Figure 19B shown, it is calculated based on the distance from the surrounding objects, that is, the distance L1 from the following vehicle M3, the distance L2 from the oncoming vehicle M2, the distance L3 from the obstacle, and the case where the size of the area that cannot be detected by the sensor of the host vehicle M1 is small, etc.

[0078] [Effect of Embodiment Example 2] According to the vehicle control device 1 of this embodiment example, even when the driving is replaced, the distance and safety from surrounding objects can be ensured to a certain extent, and the possibility of causing a sense of insecurity to the driving replacer can be reduced. In particular, in this embodiment example, when there is a dead zone area in the external sensor, a driving replacement position that is a possible position for mode transfer is set, so that an appropriate driving replacement position considering the dead zone area of the external sensor can be set. In addition, in this embodiment example, even if a position where the driving risk level of the vehicle M1 determined according to the dead zone area of the external sensor or the presence or absence of an obstacle is below the threshold is set as the driving replacement position, an appropriate driving replacement position can be set.

[0079] <Embodiment Example 3> Next, with reference to FIGS. 20 to Figure 22 The vehicle control device and the vehicle control method according to Embodiment Example 3 of the present invention will be described. In Embodiment Example 3, the structure of the vehicle control device 1 and the like included in the vehicle 100 is the same as the structure described in Embodiment Example 1. In Embodiment Example 3, the process of generating the standby trajectory is different from that in Embodiment Example 1.

[0080] [Examples of dead zones of the own vehicle and standby trajectories considering the dead zones] The vehicle control device 1 of this embodiment example predicts the state of the dead zone around the future position during the execution of the cooperative action and the actions of other vehicles, and generates a standby trajectory that estimates the detection range of the sensor for detecting the surrounding objects of the vehicle. Figure 20A , Figure 20B , Figure 20C An example of the dead zone of the own vehicle is shown. Figure 20A In the example shown, there is a dead zone causing object 761 near the road boundary 741. Here, it is assumed to be a relatively high wall or a utility pole or the like. Due to this dead zone causing object 761, a dead zone area 700 of the sensor 2 is generated near the avoidance space.

[0081] In this Figure 20A example, although the avoidance position 702 has been generated, an obstacle 762 may enter the dead zone area 700 near it and hinder the passage of the vehicle M1. In this case, as Figure 20B shown, an object (obstacle 762) that has entered the front detection range 2a of the sensor 2 for detecting in front of the vehicle M1 can be detected.

[0082] On the other hand, in the case of using the standby trajectory to reverse to the driving replacement position at the time of abandonment, the object entering the rear detection range 5a can be detected by the sensor 5 behind the own vehicle M1. However, depending on the type of sensor mounted on the vehicle M1, the detectable range is different. In addition, in the case where an optical camera or the like is used as the recognition sensor, there may be a difference in the detection ability in front of and behind the vehicle due to the influence of the surrounding lighting environment such as street lights and the headlights of the surrounding vehicles.

[0083] Therefore, for example, for the obstacle 762 that can be detected when advancing toward the oncoming vehicle position, consider the case where it cannot be detected and the case where the detection is insufficient when retreating to the driving replacement position at the time of abandonment. Specifically, as Figure 20C shown, assume the following scenario: In the rear detection range 5a of the sensor 5 that detects behind the own vehicle M1, the obstacle 762 cannot be detected.

[0084] In consideration of this situation, the standby trajectory generation unit 245 of the vehicle control device 1 of the present embodiment example generates a standby trajectory based on each point on the oncoming vehicle trajectory that predicts the detection state of the own vehicle sensor in the future.

[0085] Figure 21A The example of

[0086] Figure 21B shows the following situation: When the own vehicle M1 is traveling on the cooperative action trajectory 704 to the waiting avoidance position 702, an obstacle 762 in the dead angle area 700 is detected by the front recognition sensor on the way. Here, the vehicle control device 1 confirms whether the obstacle 762 detected by the front recognition sensor can be detected from the future position 771 using the rear recognition sensor. As a method thereof, numerical modeling of the surrounding environment such as walls around the own vehicle, other vehicles, and the sensors of the own vehicle is performed. Then, the standby trajectory generation unit 245 calculates whether a standby trajectory can be generated from the future position 771 through an operation using the numerical model.

[0087] [Generation process of standby trajectory considering the dead angle of the own vehicle] Figure 22 is a flowchart showing the process in which the standby trajectory generation unit 245 generates a standby trajectory in the present embodiment example. First, the standby trajectory generation unit 245 acquires various information (step S61). Next, the standby trajectory generation unit 245 performs numerical modeling on the host vehicle and surrounding objects such as mounted sensors, falling objects, walls, and traffic participants, and oncoming vehicles that are objects of cooperative action (step S62). When numerically modeling the vehicle, the standby trajectory generation unit 245 uses a two-wheeled vehicle model or a four-wheeled vehicle model to model the behavior of the vehicle. In addition, regarding the shape of the vehicle, it is also modeled based on the shape and color information of the vehicle detected by a camera or the like, and the size information of the vehicle based on point group information such as lidar.

[0088] Next, the driving replacement position generation unit 244 predicts a future abandonment position of the cooperative action trajectory that is the passing trajectory (step S63). Then, the standby trajectory generation unit 245 generates a future standby trajectory candidate from the future position (abandonment position) of the cooperative action trajectory predicted by the driving replacement position generation unit 244 to the driver replacement position (step S64). Thus, by generating the future standby trajectory candidate, the standby trajectory 722 is updated. As the driver replacement position, the cooperative action start position or the search result described in the second embodiment is used. In addition, although the standby trajectory 772 is updated here, the trajectory 704 (first driving trajectory) from the current position to the avoidance position 702 may also be updated.

[0089] After that, the cooperative action trajectory selection unit 246 evaluates the generated future standby trajectory candidates and selects the best trajectory (step S65). Then, the cooperative action trajectory selection unit 246 confirms whether a future standby trajectory has been generated (step S66). At the time of this confirmation, as Figure 21B shown, when the vehicle M1 is traveling on the trajectory 704 (first driving trajectory) from the current position to the avoidance position 702, the difference between the detection result of the external sensor (the front recognition sensor 2 shown in FIG. 20) and the detection result of the external sensor (the rear recognition sensor 5 shown in FIG. 20) when the vehicle is traveling on the standby trajectory 772 (second driving trajectory) is obtained or estimated, and based on this, it is determined whether an appropriate standby trajectory 772 has been obtained.

[0090] That is, when there are differences in the detection range or detection performance among multiple external sensors, the cooperative action trajectory selection unit 246 takes the difference in the detection results into consideration to determine whether an appropriate future standby trajectory has been generated. Specifically, as Figure 20C shown, when an obstacle 762 is detected by the front recognition sensor, even if the obstacle 762 cannot be detected by the rear recognition sensor as Figure 20C shown during reverse travel, the recognition result that there is an obstacle 762 is corrected based on the result detected by the front recognition sensor, and a standby trajectory 772 that avoids the obstacle 762 is obtained.

[0091] In step S66, when a suitable future standby trajectory is generated based on obtaining or estimating the difference in the detection results of the two sensors 2 and 5 (step S66 is YES), the cooperative action trajectory selection unit 246 stores the future standby trajectory information selected in step S65 in the storage unit 308 and ends the process (step S67). In addition, when it is determined to be YES in step S66, in addition to being able to generate a future standby trajectory, it also includes the case where the trajectory 704 from the current position to the position to be avoided 702 is updated.

[0092] In addition, in step S66, when a suitable future standby trajectory cannot be generated based on obtaining or estimating the difference in the detection results of the two sensors 2 and 5 (step S66 is NO), the cooperative action trajectory selection unit 246 makes a transfer request to the manual driving mode (step S68). That is, in a situation where a future standby trajectory cannot be generated, if the cooperative action is directly continued, a standby trajectory cannot be generated when a stalemate occurs at a future position, and it may be abandoned at the place where the stalemate occurs, etc. Therefore, before the stalemate occurs, a request to the manual driving mode is executed. Alternatively, it is possible to move to a driving replacement position using the standby trajectory that can be generated at the current moment instead of making a request to the manual driving mode.

[0093] [Effect of Embodiment Example 3] As described above, according to this embodiment example, when passing by another vehicle while in a state where there is a blind spot area, by generating a trajectory that takes into account the standby trajectory from a future position, it is possible to avoid the attitude or target trajectory of the position to be avoided being changed during the passing-by driving. In addition, even in a situation where abandonment is required, it does not affect the riding comfort and does not give passengers a sense of insecurity. In addition, even when there are differences in the performance (detection range, detection performance, etc.) of multiple external sensors mounted on the vehicle, it is possible to identify obstacles, etc. based on the difference in the detection results of each sensor, correct the recognition result, and generate a suitable standby trajectory, thereby generating a suitable standby trajectory that takes into account the performance differences of the external sensors.

[0094] <Variants of Each Embodiment Example> In addition, each of the embodiment examples described here is a detailed description for facilitating the understanding of the present invention and is not limited to having all the structures described. In addition, the structures and processes described in the above embodiment examples can be variously deformed and changed. For example, the passing-by operation described in FIG. 5 and the like is an example in which the own vehicle approaches the left side of the road in an area where traffic moves on the left side of the road. However, in an area where traffic moves on the right side of the road, the left and right directions are reversed when passing by another vehicle.

[0095] In addition, in each of the above-described embodiments, a driving trajectory (a first driving trajectory and a second driving trajectory) is generated (calculated) and the vehicle travels on this driving trajectory by autonomous driving. Generally, when referring to a driving trajectory, it includes information on the coordinate positions on the road and the time when passing through these coordinate positions. In the case of the present invention, the information on the coordinate positions and the time is also used to control the driving. On the other hand, as the driving trajectory, it is also possible to set it as information of a so-called driving route that does not have information on the time when the vehicle passes through each coordinate position but only has information on the coordinate positions on the road. Preferably, on the basis of performing oncoming vehicle driving, the position of the oncoming vehicle at each moment is predicted, and it is determined to which position the own vehicle moves at each moment. However, in a situation where the oncoming vehicle stops or the like, a driving route having only information on the coordinate positions can be generated in place of the driving trajectory. By generating such a driving route in this way, it is not necessary to process time information, and the driving trajectory (driving route) can be generated more simply.

[0096] In addition, Figure 2 In the structure shown, the vehicle control device 1 constituted by a computer is configured as a device for performing the processing of each of the above-described embodiments. However, the same processing can also be performed by modifying the program installed in the existing vehicle control device 1. Regarding the program in this case, in addition to preparing it in the memory in the computer shown in Figure 2 it can also be placed in an external memory, an IC card, an SD card, an optical disc or other storage media and transmitted. In addition, part or all of the vehicle control device 1 can be implemented by dedicated hardware such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).

[0097] In addition, Figure 2 , Figure 3 , Figure 6 , Figure 7 , Figure 17 In the structural diagrams shown, control lines and information lines necessary for explanation are shown, but it is not limited to showing all the control lines and information lines necessary for the product. In fact, it can be considered that almost all the structures are connected to each other. In addition, for Figure 4 , Figure 8 , Figure 10 , Figure 15 , Figure 16 , Figure 22 shown in the flowchart, if the processing results are the same, the processing order can also be changed, or multiple processes can be executed simultaneously. Description of Reference Numerals

[0098] 1 Vehicle control device 2 Front recognition sensor 3 Left-side recognition sensor 4 Right-side recognition sensor 5 Rear recognition sensor 6 Steering wheel 7 Steering torque detection device 8 Steering control device 9 Master cylinder 10 Steering control mechanism 11 FL wheel speed sensor 12 Brake pedal 13 Brake control mechanism 14 Composite sensor 15 Brake control device 16 FL, 16 FR, 16 RL, 16 RR wheel cylinders 17 Accelerator pedal 18 Stroke sensor 19 Acceleration control device 20 Throttle control mechanism 21 Steering wheel angle detection device 23 Communication device 24 Display device 100 Vehicle 101 FL Left front wheel 101 FR Right front wheel 101 RL Left rear wheel 101 RR Right rear wheel 201 Autopilot planning department 202 Automatic parking planning department 203 Vehicle motion control department 204 Actuator control department 205 Retention risk map generation department 206 Vehicle network 211 CPU 212 Memory 213 Input / output unit 214 Interface 221 Autopilot plan generation department 222 Driving mode management department 230 Trajectory planning department 231 Lane keeping trajectory generation department 232 Lane change trajectory generation department 233 Obstacle avoidance trajectory generation unit 234 Cooperative action trajectory generation unit 235 Trajectory mediation unit 241 Cooperative action status management unit 242 Standby position and attitude generation unit 243 Cooperative action trajectory calculation unit 244 Driving replacement position generation unit 245 Backup trajectory generation unit 246 Cooperative action trajectory selection unit 251 Manual driving mode 252 System shutdown mode 253 Autopilot mode 301 Radar 302 Stereo camera 303 Vehicle sensor 304 Lidar 305 Sensor information processing unit 306 Map information processing unit 307 Moving object action prediction unit 308 Storage unit 309 Residence risk map calculation unit 310 Own position estimation processing unit M1 This vehicle M2 Oncoming vehicle M3 Following vehicle.

Claims

1. A vehicle control device controls a vehicle that can transfer between an autonomous driving mode in which at least part or all of the driving tasks are implemented in a vehicle system and a manual driving mode in which a driver or a remote monitor implements the driving tasks. The vehicle control device is characterized by including: An identification unit that identifies the surrounding conditions of the vehicle based on the detection results of external sensors provided in the vehicle; A travel trajectory generation unit that generates a travel trajectory from the current position of the vehicle to a target position based on the surrounding conditions of the vehicle identified by the identification unit; And A transferable position determination unit that determines a transferable position at which the vehicle can transfer from the autonomous driving mode to the manual driving mode during the implementation of the driving task in the autonomous driving mode, The travel trajectory generation unit generates a travel trajectory of the vehicle from the current position to the target position, i.e., a first travel trajectory, and a travel trajectory from a point on the first travel trajectory to the transferable position, i.e., a second travel trajectory, The vehicle is controlled using the first travel trajectory or the second travel trajectory.

2. The vehicle control device according to claim 1, characterized in that The first travel trajectory and the second travel trajectory are information on coordinate positions on a road and do not have information on the time at the coordinate positions.

3. The vehicle control device according to claim 1, characterized in that When it is determined that it is difficult to continue controlling the vehicle in the autonomous driving mode on the first travel trajectory, the second travel trajectory is used to drive the vehicle to the transferable position, The vehicle transfers from the autonomous driving mode to the manual driving mode at the transferable position.

4. The vehicle control device according to claim 1, characterized in that The transferable position determination unit determines either the start position of the driving task in the autonomous driving mode or a position at which the driving risk of the vehicle determined by the dead angle area of the external sensor and the presence or absence of obstacles is below a threshold value as the transferable position.

5. The vehicle control device according to claim 1, characterized in that The travel trajectory generation unit obtains or estimates the difference between the detection results of the external sensors when the vehicle travels on the first travel trajectory and the detection results of the external sensors when the vehicle travels on the second travel trajectory, and updates the first travel trajectory or the second travel trajectory when the difference is above a threshold value.

6. The vehicle control device according to claim 1, characterized in that When generating the second travel trajectory, the travel trajectory generation unit corrects the recognition result according to the difference in the detection results of multiple external sensors provided in the vehicle to obtain a suitable travel trajectory.

7. A vehicle control method for controlling a vehicle that can transfer between an autonomous driving mode in which at least part or all of the driving tasks are implemented in a vehicle system and a manual driving mode in which a driver or a remote monitor implements the driving tasks, the vehicle control method being characterized by including: An identification process that identifies the situation around the vehicle based on the detection results of external sensors provided in the vehicle; A travel trajectory generation process that generates a travel trajectory from the current position of the vehicle to a target position based on the situation around the vehicle identified by the identification process; And A transferable position determination process that determines a transferable position at which a transfer from the autonomous driving mode to the manual driving mode can be made during the implementation of the driving task in the autonomous driving mode, In the travel trajectory generation process, a travel trajectory of the vehicle from the current position to the target position, i.e., a first travel trajectory, and a travel trajectory from a point on the first travel trajectory to the transferable position, i.e., a second travel trajectory, are generated respectively, and the vehicle is controlled using the generated first travel trajectory or second travel trajectory.

Images