Autonomous Driving System, Autonomous Driving Control Method, and Non-Transitory Recording Medium

Abstract

The present invention relates to an autonomous driving system, an autonomous driving control method, and a non-transitory recording medium. The autonomous driving system includes: one or more storage devices configured to store specific location information indicating specific locations where remote support may be required; and a processor configured to: determine whether there is an abnormality in a remote support system configured to provide the remote support to an autonomous driving vehicle; in a case where the abnormality in the remote support system is detected, based on the specific location information, set any of the specific locations on a target route from a current position of the autonomous driving vehicle to a destination as a boundary position; set a target evacuation position such that the target evacuation position is included in the target route from the current position to the boundary position; and control the autonomous driving vehicle to stop at the target evacuation position.

Description

Technical Field

[0001] The present disclosure relates to an autonomous driving system, an autonomous driving control method, and a non-transitory recording medium.

[0002] This application claims the priority of Japanese Patent Application No. 2021-080494 filed on May 11, 2021, and the entire contents thereof are incorporated herein by reference. Background Art

[0003] Japanese Unexamined Patent Application Publication No. 2018-077649 discloses a remote driving control device that performs remote driving of a vehicle. The remote driving control device performs remote driving of the vehicle by communicating with the vehicle.

[0004] Consider a remote support technology for remotely supporting the driving of an autonomous driving vehicle. Remote support requires communication between a remote support device and an autonomous driving vehicle. The "remote support system" includes components and functions for providing remote support to an autonomous driving vehicle. For example, the remote support system includes a remote support device, a communication network, a communication device mounted on the autonomous driving vehicle, etc. When an abnormality occurs in at least a part of the remote support system, remote support cannot be provided to the autonomous driving vehicle, or the accuracy of the remote support will decrease. Summary of the Invention

[0005] The present disclosure provides a technology capable of appropriately controlling an autonomous driving vehicle when an abnormality occurs in a remote support system that provides remote support to the autonomous driving vehicle.

[0006] The first aspect of the present disclosure is related to an autonomous driving system configured to control an autonomous driving vehicle that is an object of remote support.

[0007] The autonomous driving system includes: one or more storage devices configured to store specific location information indicating specific locations where remote support may be required; and one or more processors configured to: determine whether there is an abnormality in a remote support system configured to provide remote support to the autonomous driving vehicle; in the case where an abnormality in the remote support system is detected, based on the specific location information, set an arbitrary specific location on a target route from the current position of the autonomous driving vehicle to the destination as a boundary position; set a target evacuation position such that the target evacuation position is included in the target route from the current position to the boundary position; and control the autonomous driving vehicle to stop at the target evacuation position.

[0008] In the above first solution, alternatively, the one or more processors may also obtain a first specific position on the target route that is closest to the current position based on the specific position information, and set the first specific position as the boundary position.

[0009] In the above first solution, alternatively, the one or more processors may also obtain a first specific position on the target route that is closest to the current position based on the specific position information, determine whether the first specific position satisfies an admissibility condition, and if the first specific position does not satisfy the admissibility condition, obtain a second specific position on the target route that is different from the first specific position based on the specific position information, and set the second specific position as the boundary position. The admissibility condition may include at least one of (i) the distance between the current position and the first specific position is equal to or greater than a distance threshold, and (ii) the vehicle control amount required to stop the autonomous vehicle in front of the first specific position is equal to or less than a control amount threshold.

[0010] In the above first solution, alternatively, the one or more processors may also obtain a first specific position on the target route that is closest to the current position based on the specific position information, and set the first specific position as the boundary position when the abnormality of the remote support system is a malfunction.

[0011] In the above first solution, alternatively, the one or more processors may also obtain a first specific position on the target route that is closest to the current position based on the specific position information, determine whether the first specific position satisfies an admissibility condition when the abnormality of the remote support system is a performance degradation, and if the first specific position does not satisfy the admissibility condition, obtain a second specific position on the target route that is different from the first specific position based on the specific position information, and set the second specific position as the boundary position. The admissibility condition may include at least one of (i) the distance between the current position and the first specific position is equal to or greater than a distance threshold, and (ii) the vehicle control amount required to stop the autonomous vehicle in front of the first specific position is equal to or less than a control amount threshold.

[0012] The second solution of the present disclosure is related to an autonomous driving control method, which is executed by one or more processors to control an autonomous vehicle that is the object of remote support.

[0013] The automatic driving control method includes: determining whether there is an abnormality in the remote support system, where the remote support system provides remote support for the autonomous vehicle; in the case of detecting an abnormality in the remote support system, based on specific location information, setting an arbitrary specific location on the target route from the current position of the autonomous vehicle to the destination as the boundary position; setting a target evacuation position such that the target evacuation position is included in the target route from the current position to the boundary position; and controlling the autonomous vehicle to stop at the target evacuation position. The specific location information indicates specific locations where remote support may be required.

[0014] The third aspect of the present disclosure is associated with a non-transitory recording medium that stores a command that can be executed by a computer and causes the computer to perform a function of controlling an autonomous vehicle that is the object of remote support.

[0015] The function includes: determining whether there is an abnormality in the remote support system, where the remote support system provides remote support for the autonomous vehicle; in the case of detecting an abnormality in the remote support system, based on specific location information, setting an arbitrary specific location on the target route from the current position of the autonomous vehicle to the destination as the boundary position; setting a target evacuation position such that the target evacuation position is included in the target route from the current position to the boundary position; and controlling the autonomous vehicle to stop at the target evacuation position. The specific location information indicates specific locations where remote support may be required.

[0016] According to the aspects of the present disclosure, in the case of detecting an abnormality in the remote support system, the target evacuation position is set considering specific locations where remote support may be required. Specifically, an arbitrary specific location on the target route to the destination is set as the boundary position. And, the target evacuation position is set to be included in the target route from the current position of the autonomous vehicle to the boundary position.

[0017] The target evacuation position does not need to be near the current position of the autonomous vehicle, and can be in front of the boundary position. Therefore, the target evacuation position can be set in a way that the autonomous vehicle can park with sufficient space. According to the aspects of the present disclosure, the safety of the autonomous vehicle and surrounding vehicles can be improved.

[0018] In addition, it is not necessary to end the automatic driving near the current position, and the automatic driving can continue to the target evacuation position. According to the aspects of the present disclosure, the continuity of the automatic driving can be improved.

[0019] Moreover, any specific position on the target route is set as a boundary position. Therefore, when compared with the case where the autonomous driving vehicle has to travel to the destination, the number of specific positions passed by the autonomous driving vehicle is reduced. Since the number of specific positions passed by the autonomous driving vehicle is reduced, the probability of requiring remote support decreases as a whole. Thus, the impact caused by the abnormality of the remote support system is at least reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Hereinafter, with reference to the drawings, the features, advantages, and technical and industrial significance of the exemplary embodiments of the present invention will be described, where the same reference numerals denote the same elements, and:

[0021] Figure 1 is a conceptual diagram showing a remote support system according to an embodiment of the present disclosure.

[0022] Figure 2 is a conceptual diagram for explaining an overview of remote support according to an embodiment of the present disclosure.

[0023] Figure 3 is a conceptual diagram for explaining an example of a specific position according to an embodiment of the present disclosure.

[0024] Figure 4 is a conceptual diagram for explaining an example of a retreat process according to an embodiment of the present disclosure.

[0025] Figure 5 is a conceptual diagram for explaining an example of a parking candidate area according to an embodiment of the present disclosure.

[0026] Figure 6 is a conceptual diagram for explaining an example of a high-priority area and a low-priority area according to an embodiment of the present disclosure.

[0027] Figure 7 is a conceptual diagram for explaining the processing when an abnormality occurs in the remote support system according to an embodiment of the present disclosure.

[0028] Figure 8 is a conceptual diagram for explaining an example of the processing when an abnormality occurs in the remote support system according to an embodiment of the present disclosure.

[0029] Figure 9 is a conceptual diagram for explaining another example of the processing when an abnormality occurs in the remote support system according to an embodiment of the present disclosure.

[0030] Figure 10 is a block diagram showing a configuration example of an autonomous driving system according to an embodiment of the present disclosure.

[0031] Figure 11It is a block diagram showing an example of driving environment information according to an embodiment of the present disclosure.

[0032] Figure 12 It is a flowchart showing an example of processing implemented by an autonomous driving system according to an embodiment of the present disclosure.

[0033] Figure 13 It is a flowchart showing a first example of step S300 according to an embodiment of the present disclosure.

[0034] Figure 14 It is a flowchart showing a second example of step S300 according to an embodiment of the present disclosure.

[0035] Figure 15 It is a flowchart showing a third example of step S300 according to an embodiment of the present disclosure. Detailed Embodiments

[0036] Embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0037] 1. Outline of Remote Support

[0038] Figure 1 It is a conceptual diagram showing a remote support system according to the present embodiment. The remote support system includes an autonomous driving vehicle 1, a remote support device 2, and a communication network 3.

[0039] The autonomous driving vehicle 1 is a vehicle capable of autonomous driving. As the autonomous driving here, autonomous driving on the premise that the driver does not have to be 100% concentrated on driving (so-called level 3 or higher autonomous driving) is assumed. The autonomous driving vehicle 1 may also be an autonomous driving vehicle of level 4 or higher that does not require a driver. The autonomous driving vehicle 1 is the object of remote support in the present embodiment.

[0040] The remote support device 2 is a device for remotely supporting the autonomous driving vehicle 1 and is operated by a remote operator. The autonomous driving vehicle 1 and the remote support device 2 are connected via the communication network 3 so as to be able to communicate with each other. The remote support device 2 communicates with the autonomous driving vehicle 1 via the communication network 3 and remotely supports the driving of the autonomous driving vehicle 1. More specifically, the remote operator operates the remote support device 2 to remotely support the driving of the autonomous driving vehicle 1. The remote support device 2 can also be said to be a device that assists the remote operator in remotely supporting the autonomous driving vehicle 1.

[0041] The communication network 3 includes a wireless base station, a wireless communication network, a wired communication network, etc. As the wireless communication network, for example, a 5G (5th Generation Mobile Communication Technology) network is exemplified.

[0042] Figure 2 This is a conceptual diagram for explaining the overview of remote support for this embodiment. The autonomous driving system 10 controls the autonomous driving vehicle 1. During autonomous driving, the autonomous driving system 10 performs various vehicle processes. As representative vehicle processes during autonomous driving, the following processes can be cited.

[0043] (1) Recognition process: The autonomous driving system 10 uses recognition sensors to recognize the situation around the autonomous driving vehicle 1. For example, the autonomous driving system 10 uses a camera to recognize the signal display of traffic lights (e.g., green light, yellow light, red light, right turn signal, etc.).

[0044] (2) Action judgment process: The autonomous driving system 10 judges whether to perform an action based on the result of the recognition process. As actions, starting, stopping, turning right, turning left, lane change, etc. are exemplified.

[0045] (3) Timing judgment process: The autonomous driving system 10 judges the execution timing of performing the above actions.

[0046] Typically, the situation that requires remote support by a remote operator is a situation where it is difficult to perform autonomous driving. For example, at Figure 3 an intersection as shown, remote support may be required.

[0047] For example, when sunlight shines on the traffic lights installed at an intersection, the recognition accuracy of the signal display may decrease. When the signal display cannot be accurately discriminated through the recognition process, the autonomous driving system 10 requires remote support for signal recognition. In addition, when the signal display cannot be discriminated, it is also difficult to judge at which timing which action should be performed. Thus, the autonomous driving system 10 also requires remote support for the action judgment process and the timing judgment process.

[0048] It is also possible to consider a situation where it is difficult to judge whether an action can actually be performed even if the signal display is discriminated. For example, although the signal display observed from the autonomous driving system 10 becomes "can turn right", sometimes an oncoming vehicle may enter the intersection, or an oncoming vehicle or a preceding vehicle may be stuck in the intersection. In such a case, the autonomous driving system 10 can also request remote support for the action judgment process and the timing judgment process while maintaining a stopped state.

[0049] As another example, it is also possible to consider a situation where it is difficult to judge whether to perform a lane change when there is a construction section in front of the autonomous driving vehicle 1. In this case, the autonomous driving system 10 can also request remote support for the action judgment process.

[0050] The autonomous driving system 10 can also request remote driving (remote operation) of the autonomous driving vehicle 1 from a remote operator. "Remote support" in this embodiment refers to the concept that includes not only at least one type of support in recognition processing, action determination processing, and timing determination processing, but also remote driving (remote operation).

[0051] When it is determined that remote support is required, the autonomous driving system 10 sends a remote support request REQ to the remote support device 2 via the communication network 3. The remote support request REQ is information that requests remote support for the autonomous driving vehicle 1 from the remote operator. The remote support device 2 notifies the received remote support request REQ to the remote operator. In response to the remote support request REQ, the remote operator starts remote support for the autonomous driving vehicle 1.

[0052] During the process of remote support, the autonomous driving system 10 sends vehicle information VCL to the remote support device 2 via the communication network 3. The vehicle information VCL represents the state of the autonomous driving vehicle 1, the surrounding conditions, the results of vehicle processing performed by the autonomous driving system 10, and so on. The remote support device 2 presents the vehicle information VCL received from the autonomous driving system 10 to the remote operator. For example, as Figure 2 shown, the remote support device 2 displays the image information IMG captured by the camera mounted on the autonomous driving vehicle 1 on the display device.

[0053] The remote operator performs remote support for the autonomous driving vehicle 1 while referring to the vehicle information VCL. The operator instruction INS is an instruction for the autonomous driving vehicle 1 input by the remote operator. The remote support device 2 receives the input of the operator instruction INS from the remote operator. Then, the remote support device 2 sends the operator instruction INS to the autonomous driving vehicle 1 via the communication network 3. The autonomous driving system 10 receives the operator instruction INS from the remote support device 2 and controls the autonomous driving vehicle 1 according to the received operator instruction INS.

[0054] 2. Processing When an Abnormality Occurs in the Remote Support System

[0055] 2-1. Abnormality of the Remote Support System

[0056] In this embodiment, the "remote support system 4" refers to the components and functions for providing remote support to the autonomous driving vehicle 1. For example, the remote support system 4 includes the remote support device 2, the communication network 3, the communication device mounted on the autonomous driving vehicle 1, etc. (refer to Figure 1 ). As the communication device mounted on the autonomous driving vehicle 1, examples include a communication ECU (Electronic Control Unit), a communication module, a transceiver circuit, etc.

[0057] Hereinafter, consider a case where an "abnormality" has occurred in at least a part of the remote support system 4 that provides remote support to the autonomous driving vehicle 1.

[0058] For example, an abnormality of the remote support system 4 includes a "functional failure" in which the functions of the remote support system 4 are lost. An example of a functional failure of the remote support system 4 is a communication interruption. For example, a communication interruption may occur when there is a problem in the communication network 3. Another example of a functional failure of the remote support system 4 is a failure (shutdown) of the remote support device 2.

[0059] Another example of a functional failure of the remote support system 4 is a failure of the communication device mounted on the autonomous driving vehicle 1. When a functional failure of the remote support system 4 occurs, remote support cannot be provided to the autonomous driving vehicle 1.

[0060] An abnormality of the remote support system 4 may also include a "performance degradation" in which the functions of the remote support system 4 decline. An example of a performance degradation of the remote support system 4 is a significant decrease in communication speed and throughput. Another example of a performance degradation of the remote support system 4 is a significant increase in communication latency. Another example of a performance degradation of the remote support system 4 is a decrease in the internal communication speed and calculation speed in the communication ECU mounted on the autonomous driving vehicle 1. When a performance degradation of the remote support system 4 occurs, the accuracy of remote support may decline.

[0061] 2-2. Retraction processing

[0062] When an abnormality of the remote support system 4 occurs, remote support cannot be provided to the autonomous driving vehicle 1, or the accuracy of remote support may decline. Therefore, when an abnormality of the remote support system 4 is detected, the autonomous driving system 10 performs a "retraction processing" that safely retracts the autonomous driving vehicle 1.

[0063] Figure 4 is a conceptual diagram for explaining an example of the retraction processing of the present embodiment. The "target retraction position PE" is the target stop position when the autonomous driving vehicle 1 stops by the retraction processing. The target retraction position PE can be set at a safe position on the road. In Figure 4In the example shown, the target avoidance position PE is set on the road shoulder. The autonomous driving system 10 controls the autonomous driving vehicle 1 to travel toward the target avoidance position PE and stop at the target avoidance position PE. For example, the autonomous driving system 10 generates a target trajectory TR in which the autonomous driving vehicle 1 travels from the current position toward the target avoidance position PE and stops at the target avoidance position PE. Then, the autonomous driving system 10 controls the driving of the autonomous driving vehicle 1 so that the autonomous driving vehicle 1 follows the target trajectory TR.

[0064] It is also possible to pre-specify an area that can be used as the target avoidance position PE in the avoidance process. Hereinafter, the area that can be used as the target avoidance position PE in the avoidance process is referred to as the "parking candidate area AC".

[0065] Figure 5 It is a conceptual diagram for explaining an example of the parking candidate area AC. In order to explain the parking candidate area AC, first, the "parking prohibited area AX" is explained. The parking prohibited area AX is an area where vehicle parking is prohibited, and is pre-specified by the Road Traffic Law, etc. In Figure 5 In the example shown, the parking prohibited area AX includes a crosswalk and a prescribed-width area around it. The parking prohibited area AX may also include a crossroads and a prescribed-width area around it. In addition, an area in front of a fire-fighting facility, etc. is also included in the parking prohibited area AX.

[0066] The parking candidate area AC is selected from an area outside the parking prohibited area AX on the road. Typically, the parking candidate area AC is a part of an area outside the parking prohibited area AX. For example, the parking candidate area AC is selected from the viewpoint of ensuring the safety of the autonomous driving vehicle 1 after parking. As Figure 5 illustrated, the parking candidate area AC may also be an area relatively close to the road end. The parking candidate area AC may also be set to include a road shoulder and a roadside strip.

[0067] As Figure 6 shown, it is also possible to set a priority for the parking candidate area AC. The high-priority area ACH is a parking candidate area AC with a higher priority, and the low-priority area ACL is a parking candidate area AC with a lower priority. In Figure 6 In the example shown, a straight section is set as the high-priority area ACH, and a curved section is set as the low-priority area ACL.

[0068] The parking candidate area AC and the parking prohibited area AX are registered in the map information in advance, for example. In the avoidance process, the autonomous driving system 10 may also set the target avoidance position PE to be included in the parking candidate area AC. When priorities are set for the parking candidate areas AC, the autonomous driving system 10 sets the target avoidance position PE to be included in the parking candidate area AC with the highest possible priority.

[0069] 2-3. Avoidance margin

[0070] As described above, when an abnormality of the remote support system 4 is detected, the autonomous driving system 10 performs an avoidance process. However, it is not necessarily required to immediately stop the autonomous driving vehicle 1 immediately after detecting the abnormality of the remote support system 4. This is because, in a situation where remote support is not required, the autonomous driving system 10 can continue with autonomous driving as usual. That is, even though an abnormality of the remote support system 4 is detected, there is no need to urgently perform a forced lane change or sudden deceleration. According to the present embodiment, the autonomous driving system 10 also sets the target avoidance position PE in consideration of the possibility of requiring remote support.

[0071] Hereinafter, a position where remote support for the autonomous driving vehicle 1 may be required will be referred to as a "specific position PS". For example, the specific position PS is Figure 3 an intersection as shown. As another example, the specific position PS may also be a position outside the ODD (Operational Design Domain, the area where autonomous driving can be performed). As still another example, the specific position PS may also include a construction section, a congested section, an accident occurrence position, etc. Typically, the specific position PS is registered in the map information in advance. Alternatively, information on the specific position PS such as a congested section or an accident occurrence position may be obtained in real time.

[0072] Figure 7 An example of the state of autonomous driving at the timing when an abnormality of the remote support system 4 is detected is shown. The current position and the destination of the autonomous driving vehicle 1 are represented by the reference numerals "P1" and "DST", respectively. The target route RT from the current position P1 of the autonomous driving vehicle 1 to the destination DST is set by the autonomous driving system 10. The autonomous driving system 10 controls the autonomous driving vehicle 1 to travel along the target route RT to the destination DST.

[0073] At the time point when an abnormality of the remote support system 4 is detected, there is a specific position PS on the target route RT to the destination DST. At Figure 7In the example shown, there are multiple specific positions PS1, PS2, and PS3 on the target route RT. The autonomous driving system 10 sets any specific position PS on the target route RT as the "limit position PL". Hereinafter, the section of the target route RT from the current position P1 to the limit position PL is referred to as the "evacuation margin section XE". The autonomous driving system 10 selects a target evacuation position PE from the evacuation margin section XE. That is, the autonomous driving system 10 sets the target evacuation position PE to be included in the evacuation margin section XE.

[0074] Figure 8 It is a conceptual diagram for explaining an example of the limit position PL and the evacuation margin section XE. In Figure 8 the example shown, the limit position PL is the first specific position PS1 closest to the current position P1 on the target route RT. The evacuation margin section XE is the section from the current position P1 to the first specific position PS1. The target evacuation position PE is selected from this evacuation margin section XE. The target evacuation position PE does not need to be near the current position P1 and can be in front of the first specific position PS1 (limit position PL). Therefore, the target evacuation position PE can be set in such a way that the autonomous driving vehicle 1 can park with a margin. In other words, the evacuation process can be carried out with a margin. Thereby, the safety of the autonomous driving vehicle 1 and the surrounding vehicles can be improved.

[0075] In addition, in Figure 8 the example shown, the autonomous driving vehicle 1 also stops at the target evacuation position PE without passing through any arbitrary specific position PS. Thus, a situation requiring remote support does not occur. Thereby, a situation where remote support is needed but not available can be avoided.

[0076] Figure 9 It is a conceptual diagram for explaining another example of the limit position PL and the evacuation margin section XE. The limit position PL is not limited to the first specific position PS1 closest to the current position P1. For example, when the current position P1 is immediately in front of the first specific position PS1, the second specific position PS2 after the first specific position PS1 can also be set as the limit position PL. The evacuation margin section XE is the section from the current position P1 to the second specific position PS2. The target evacuation position PE is selected from this evacuation margin section XE. Thereby, there is no need to make a forced lane change or sudden deceleration to stop in front of the first specific position PS1. That is, the evacuation process can be carried out with a margin.

[0077] In Figure 9In the example shown, the autonomous driving vehicle 1 needs to pass through the first specific position PS1. The first specific position PS1 is a position where remote support may be required, but remote support is not necessarily required at the first specific position PS1. If remote support is not required at the first specific position PS1, the autonomous driving vehicle 1 can pass through the first specific position PS1 using normal autonomous driving. Even assuming that remote support is required at the first specific position PS1, if the abnormality of the remote support system 4 is "performance degradation", remote support can be performed although it is slow. In this case, the autonomous driving vehicle 1 can also pass through the first specific position PS1. After that, the autonomous driving vehicle 1 stops at the target evacuation position PE before passing through the second specific position PS2. When compared with the case where the autonomous driving vehicle 1 has to travel to the destination DST, the number of specific positions PS passed by the autonomous driving vehicle 1 is reduced. Since the number of specific positions PS passed is reduced, the probability of requiring remote support decreases as a whole. Thus, the impact caused by the abnormality of the remote support system 4 is at least reduced.

[0078] 2-4. Effects

[0079] As described above, according to the present embodiment, when an abnormality of the remote support system 4 is detected, the target evacuation position PE is set in consideration of the specific position PS where remote support may be required. Specifically, any specific position PS on the target route RT to the destination DST is set as the boundary position PL. And the target evacuation position PE is set to be included in the target route RT (evacuation margin section XE) from the current position P1 of the autonomous driving vehicle 1 to the boundary position PL.

[0080] The target evacuation position PE does not need to be near the current position P1 and can be in front of the boundary position PL. Therefore, the target evacuation position PE can be set in such a way that the autonomous driving vehicle 1 can park with a margin. In other words, the evacuation process can be performed with a margin. Thus, the safety of the autonomous driving vehicle 1 and the surrounding vehicles can be improved.

[0081] In addition, it is not necessary to end autonomous driving near the current position P1, and autonomous driving can continue to the target evacuation position PE. Thus, the continuity of autonomous driving can be improved.

[0082] Moreover, any specific position PS on the target route RT is set as the boundary position PL. Therefore, when compared with the case where the autonomous driving vehicle 1 has to travel to the destination DST, the number of specific positions PS passed by the autonomous driving vehicle 1 is reduced. Since the number of specific positions PS passed by the autonomous driving vehicle 1 is reduced, the probability of requiring remote support decreases as a whole. Thus, the impact caused by the abnormality of the remote support system 4 is at least reduced.

[0083] The first specific position PS1 closest to the current position P1 on the target route RT can also be set as the limit position PL. In this case, the autonomous driving vehicle 1 stops at the target avoidance position PE without passing through any specific position PS. Thus, a situation requiring remote support does not occur. Therefore, a situation where remote support is needed but cannot be obtained can be avoided before it happens.

[0084] Hereinafter, the autonomous driving system 10 of the present embodiment will be described in further detail.

[0085] 3. Example of Autonomous Driving System

[0086] 3-1. Configuration Example

[0087] The autonomous driving system 10 controls the autonomous driving vehicle 1. Typically, the autonomous driving system 10 is mounted on the autonomous driving vehicle 1. Alternatively, at least a part of the autonomous driving system 10 may be configured in an external device outside the autonomous driving vehicle 1 to remotely control the autonomous driving vehicle 1. That is, the autonomous driving system 10 may be distributedly configured in the autonomous driving vehicle 1 and the external device.

[0088] Figure 10 It is a block diagram showing a configuration example of the autonomous driving system 10 of the present embodiment. The autonomous driving system 10 includes a sensor group 20, a driving device 30, a communication device 40, and a control device 100.

[0089] The sensor group 20 is mounted on the autonomous driving vehicle 1. The sensor group 20 includes a vehicle state sensor, an identification sensor, a position sensor, etc. The vehicle state sensor detects the state of the autonomous driving vehicle 1. As the vehicle state sensor, a vehicle speed sensor, a yaw rate sensor, a lateral acceleration sensor, a steering angle sensor, etc. are exemplified. The identification sensor detects the surrounding conditions of the autonomous driving vehicle 1. As the identification sensor, a camera, a lidar (Laser Imaging Detection and Ranging), a radar, etc. are exemplified. The position sensor detects the position and orientation of the autonomous driving vehicle 1. As the position sensor, a GPS (Global Positioning System) sensor is exemplified.

[0090] The driving device 30 is mounted on the autonomous driving vehicle 1. The driving device 30 includes a steering device, a driving device, and a braking device. The steering device steers the wheels. For example, the steering device includes an electric power steering (EPS) device. The driving device is a power source that generates a driving force. As the driving device, an engine, an electric motor, an in-wheel motor, etc. are exemplified. The braking device generates a braking force.

[0091] The communication device 40 communicates with the outside of the autonomous driving vehicle 1. For example, the communication device 40 communicates with the remote support device 2 via the communication network 3 (refer to Figure 1 , Figure 2 ). The communication device 40 may also communicate with the management server. The communication device 40 may also perform V2I (vehicle-to-infrastructure) communication (vehicle-road communication) with the surrounding infrastructure. The communication device 40 may also perform V2V (vehicle-to-vehicle) communication (vehicle-vehicle communication) with surrounding vehicles. The communication device 40 includes a communication ECU (Electronic Control Unit), a communication module, a transceiver circuit, etc.

[0092] The control device 100 controls the autonomous driving vehicle 1. The control device 100 includes one or more processors 110 (hereinafter, only referred to as the processor 110) and one or more storage devices 120 (hereinafter, only referred to as the storage device 120). The processor 110 performs various processes. For example, the processor 110 includes a CPU (Central Processing Unit). The storage device 120 stores various information. Examples of the storage device 120 include a volatile memory, a non-volatile memory, an HDD (Hard Disk Drive), an SSD (Solid State Drive), etc. The control device 100 may also include one or more ECUs. A part of the control device 100 may also be an information processing device outside the autonomous driving vehicle 1.

[0093] The autonomous driving control program PROG is a computer program for controlling the autonomous driving vehicle 1. Various processes performed by the control device 100 are realized by the processor 110 executing the autonomous driving control program PROG. The autonomous driving control program PROG is stored in the storage device 120. Alternatively, the autonomous driving control program PROG may also be recorded on a computer-readable recording medium.

[0094] 3-2. Driving environment information

[0095] The driving environment information 200 represents the driving environment of the autonomous driving vehicle 1. The driving environment information 200 is stored in the storage device 120.

[0096] Figure 11It is a block diagram showing an example of driving environment information 200. The driving environment information 200 includes map information 210, specific location information 220, avoidance area information 230, vehicle state information 240, surrounding condition information 250, vehicle position information 260, and distribution information 270.

[0097] The map information 210 includes a general navigation map. The map information 210 can also represent lane configurations, road shapes, etc. The map information 210 can also include location information such as signals and signs. The processor 110 obtains the map information of the required area from the map database. The map database can be stored in a specified storage device mounted on the autonomous vehicle 1 or in an external management server. In the latter case, the processor 110 communicates with the management server to obtain the required map information.

[0098] The specific location information 220 represents a specific location PS where remote support for the autonomous vehicle 1 may be required. For example, the specific location information 220 is prepared in advance. The specific location information 220 can also be included in the map information 210. The specific location information 220 can also be added in real time as described later.

[0099] The avoidance area information 230 represents the positions of the parking candidate area AC and the no-parking area AX (see Figure 5 ). The avoidance area information 230 can also represent the priority of the parking candidate area AC (see Figure 6 ). The avoidance area information 230 is prepared in advance. The avoidance area information 230 can also be included in the map information 210.

[0100] The vehicle state information 240 is information indicating the state of the autonomous vehicle 1. The processor 110 obtains the vehicle state information 240 from the vehicle state sensor.

[0101] The surrounding condition information 250 is information indicating the conditions around the autonomous vehicle 1. The processor 110 uses the recognition sensor to obtain the surrounding condition information 250. For example, the surrounding condition information 250 includes image information IMG captured by a camera. The surrounding condition information 250 also includes object information related to the objects around the autonomous vehicle 1. As objects, pedestrians, bicycles, other vehicles (preceding vehicles, parked vehicles, etc.), road structures (white lines, curbs, guardrails, walls, median strips, roadside structures, etc.), signs, obstacles, etc. are exemplified. The object information represents the relative position and relative speed of the object with respect to the autonomous vehicle 1.

[0102] The vehicle position information 260 is information indicating the position of the autonomous vehicle 1. The processor 110 obtains the vehicle position information 260 from the detection results obtained by the position sensor. In addition, the processor 110 can also obtain highly accurate vehicle position information 260 through well-known self-position estimation processing (Localization) that utilizes object information and map information 210.

[0103] The distribution information 270 includes road traffic information, construction section information, traffic control information, etc. The processor 110 receives the distribution information 270 from an information providing server or roadside infrastructure via the communication device 40.

[0104] The processor 110 can grasp construction sections, congested sections, accident occurrence locations, etc. based on the distribution information 270. In this case, the processor 110 can also add construction sections, congested sections, accident occurrence locations, etc. to the specific position information 220.

[0105] 3 - 3. Vehicle Driving Control, Autonomous Driving Control

[0106] The processor 110 executes "vehicle driving control" for controlling the driving of the autonomous vehicle 1. The vehicle driving control includes steering control, acceleration control, and deceleration control. The processor 110 executes vehicle driving control by controlling the driving device 30 (steering device, driving device, braking device). Specifically, the processor 110 executes steering control by controlling the steering device. In addition, the processor 110 executes acceleration control by controlling the driving device. In addition, the processor 110 executes deceleration control by controlling the braking device.

[0107] In addition, the processor 110 performs autonomous driving control based on the driving environment information 200. More specifically, the processor 110 sets a target route RT to the destination DST based on the map information 210 and the like. And the processor 110 performs vehicle driving control in such a way that the autonomous vehicle 1 travels along the target route RT to the destination DST based on the driving environment information 200.

[0108] More specifically, the processor 110 generates a driving plan for the autonomous vehicle 1 based on the driving environment information 200. The driving plan includes maintaining the current driving lane, changing lanes, avoiding obstacles, etc. Moreover, the processor 110 generates a target trajectory TR required for the autonomous vehicle 1 to travel according to the driving plan. The target trajectory TR includes a target position and a target speed. Then, the processor 110 executes vehicle driving control in such a way that the autonomous vehicle 1 follows the target route RT and the target trajectory TR.

[0109] 3 - 4. Processing Associated with Remote Support

[0110] During the process of autonomous driving, the processor 110 determines whether remote support by a remote operator is required. Typically, the situation where remote support by a remote operator is required is a situation where autonomous driving is difficult. For example, when it is difficult to perform at least one of the above-described recognition processing, action determination processing, and timing determination processing, the processor 110 determines that remote support by a remote operator is required.

[0111] When it is determined that remote support is required, the processor 110 sends a remote support request REQ to the remote support device 2 via the communication device 40. The remote support request REQ requests remote support for the autonomous driving vehicle 1 from the remote operator.

[0112] In addition, the processor 110 sends vehicle information VCL to the remote support device 2 via the communication device 40. The vehicle information VCL includes at least a part of the driving environment information 200. For example, the vehicle information VCL includes image information IMG captured by a camera. The vehicle information VCL may also include object information. The vehicle information VCL may also include vehicle state information 240 and vehicle position information 260. The vehicle information VCL may also include the results of recognition processing, action determination processing, and timing determination processing.

[0113] Moreover, the processor 110 receives an operator instruction INS from the remote support device 2 via the communication device 40. The operator instruction INS is an instruction for the autonomous driving vehicle 1 input by the remote operator. When the operator instruction INS is received, the processor 110 performs vehicle driving control according to the received operator instruction INS.

[0114] 4. Processing Flow When an Abnormality Occurs in the Remote Support System

[0115] Figure 12 It is a flowchart showing an example of processing performed by the autonomous driving system 10 of the present embodiment. In particular, Figure 12 It shows a processing flow associated with the occurrence of an abnormality in the remote support system 4.

[0116] 4-1. Step S100

[0117] In step S100, the processor 110 determines whether there is an abnormality in the remote support system 4 that provides remote support for the autonomous driving vehicle 1. For example, the remote support system 4 includes the remote support device 2, the communication network 3, and the communication device 40 of the autonomous driving system 10.

[0118] An abnormality of the remote support system 4 includes a "malfunction" in which the functions of the remote support system 4 are lost. For example, the processor 110 monitors the communication status with the remote support device 2 (e.g., throughput, communication speed). When the communication with the remote support device 2 is interrupted, the processor 110 determines that a malfunction has occurred in the remote support device 2 or the communication network 3. As another example, the communication device 40 (e.g., communication ECU) of the autonomous driving system 10 has a self-diagnosis function. The processor 110 can detect a malfunction of the communication device 40 through this self-diagnosis function.

[0119] An abnormality of the remote support system 4 may also include a performance degradation in which the functions of the remote support system 4 decline. For example, the processor 110 monitors the communication status with the remote support device 2 (e.g., throughput, communication speed, communication latency). When the throughput or communication speed is lower than the first threshold, the processor 110 determines that a performance degradation of the remote support system 4 has occurred. As another example, when the communication latency exceeds the second threshold, the processor 110 determines that a performance degradation of the remote support system 4 has occurred.

[0120] In the case where an abnormality of the remote support system 4 is not detected (step S100: No), the processing in this cycle ends. On the other hand, in the case where an abnormality of the remote support system 4 is detected (step S100: Yes), the processing proceeds to step S200.

[0121] 4 - 2. Step S200

[0122] In step S200, the processor 110 determines whether a specific position PS exists on the target route RT from the current position P1 to the destination DST. The target route RT is set and grasped by the processor 110. The specific position PS is obtained from the specific position information 220. Thus, the processor 110 can determine whether a specific position PS exists on the target route RT based on the specific position information 220.

[0123] In the case where a specific position PS exists on the target route RT (step S200: Yes), the processing proceeds to step S300. On the other hand, in the case where a specific position PS does not exist on the target route RT (step S200: No), the processing proceeds to step S400.

[0124] 4 - 3. Step S300

[0125] In step S300, the processor 110 sets an arbitrary specific position PS on the target route RT as the boundary position PL based on the target route RT and the specific position information 220. Then, the processing proceeds to step S500. Hereinafter, several examples of step S300 will be described.

[0126] 4-3-1. First Example

[0127] Figure 13 It is a flowchart showing the first example of step S300.

[0128] In step S310, the processor 110 acquires a first specific position PS1 that is the closest to the current position P1 on the target route RT. The current position P1 is obtained from the vehicle position information 260. The specific position PS is obtained from the specific position information 220. The processor 110 can acquire the first specific position PS1 based on the specific position information 220 and the vehicle position information 260.

[0129] In step S340, the processor 110 sets the first specific position PS1 as the limit position PL.

[0130] According to the first example, it is possible to stop the autonomous driving vehicle 1 without passing through an arbitrary specific position PS.

[0131] 4-3-2. Second Example

[0132] Figure 14 It is a flowchart showing the second example of step S300. Step S310 is the same as in the case of the first example.

[0133] In step S330, the processor 110 determines whether the first specific position PS1 satisfies the allowable condition. The allowable condition is set from the viewpoint of whether it is possible to stop the autonomous driving vehicle 1 without making a forced lane change or sudden deceleration.

[0134] For example, the allowable condition includes at least one of the following conditions (A) and (B).

[0135] Condition (A): The distance between the current position P1 and the first specific position PS1 is equal to or greater than a specified distance threshold.

[0136] Condition (B): The vehicle control amount required to stop the autonomous driving vehicle 1 in front of the first specific position PS1 is equal to or less than a control amount threshold. Here, as the vehicle control amount, deceleration and steering amount are exemplified.

[0137] The current position P1 of the autonomous driving vehicle 1 is obtained from the vehicle position information 260. The current vehicle speed of the autonomous driving vehicle 1 is obtained from the vehicle state information 240. The motion performance of the autonomous driving vehicle 1 is given in advance as information. The processor 110 determines whether the first specific position PS1 satisfies the allowable condition based on the current position P1, the first specific position PS1, the current vehicle speed, the motion performance, etc.

[0138] When the first specific position PS1 satisfies the allowable condition (step S330: Yes), the process proceeds to step S340. In step S340, the processor 110 sets the first specific position PS1 as the limit position PL.

[0139] On the other hand, when the first specific position PS1 does not satisfy the allowable condition (step S330: No), the process proceeds to step S350. In step S350, the processor 110 obtains a second specific position PS2 different from the first specific position PS1 on the target route RT based on the specific position information 220. For example, the second specific position PS2 is a specific position PS after the first specific position PS1 when observed from the current position P1 (refer to Figure 9 ). After that, the process proceeds to step S360.

[0140] In step S360, the processor 110 sets the second specific position PS2 as the limit position PL.

[0141] According to the second example, the autonomous driving vehicle 1 can stop without making a forced lane change or sudden deceleration.

[0142] 4 - 3 - 3. Third example

[0143] Figure 15 It is a flowchart showing the third example of step S300. Step S310 is the same as in the first example.

[0144] In step S320, the processor 110 determines whether the abnormality of the remote support system 4 is a malfunction or a performance degradation.

[0145] When the abnormality of the remote support system 4 is a malfunction (step S320: Yes), the process proceeds to step S340. In step S340, the processor 110 sets the first specific position PS1 as the limit position PL.

[0146] On the other hand, when the abnormality of the remote support system 4 is a performance degradation (step S320: No), the process proceeds to step S330. Steps S330 and later are the same as in the second example.

[0147] According to the third example, when the abnormality of the remote support system 4 is a performance degradation, the selection range of the target avoidance position PE can be expanded. That is, when the abnormality of the remote support system 4 is a performance degradation, the conditions imposed on the target avoidance position PE can be relaxed.

[0148] 4 - 4. Step S400

[0149] In step S400, the processor 110 sets the destination DST as the limit position PL. After that, the process proceeds to step S500.

[0150] 4 - 5. Step S500

[0151] In step S500, the processor 110 sets the target avoidance position PE based on the limit position PL. Specifically, the processor 110 acquires the section of the target route RT from the current position P1 to the limit position PL as the avoidance surplus section XE. Then, the processor 110 selects the target avoidance position PE from the avoidance surplus section XE. That is, the processor 110 sets the target avoidance position PE to be included in the avoidance surplus section XE.

[0152] It should be noted that the processor 110 sets the target avoidance position PE in such a way that the autonomous driving vehicle 1 can actually stop based on the current position P1, vehicle speed, motion performance, etc. of the autonomous driving vehicle 1.

[0153] When setting the target avoidance position PE, the processor 110 may also refer to the avoidance area information 230. The avoidance area information 230 shows the positions of the parking candidate area AC and the parking prohibited area AX. The processor 110 avoids the parking prohibited area AX and sets the target avoidance position PE in the parking candidate area AC included in the avoidance surplus section XE. The avoidance area information 230 may also show the priority of the parking candidate area AC. In this case, the processor 110 sets the target avoidance position PE to be included in the parking candidate area AC with the highest possible priority.

[0154] 4 - 6. Step S600

[0155] In step S600, the processor 110 controls the vehicle driving in such a way that the autonomous driving vehicle 1 travels toward the target avoidance position PE and stops at the target avoidance position PE. For example, the processor 110 generates a target trajectory TR in which the autonomous driving vehicle 1 travels from the current position P1 to the target avoidance position PE and stops at the target avoidance position PE. Then, the autonomous driving system 10 controls the vehicle driving in such a way that the autonomous driving vehicle 1 follows the target trajectory TR (refer to Figure 4 ).

Claims

1. An autonomous driving system configured to control an autonomous vehicle that is the object of remote support, wherein the autonomous driving system includes: One or more storage devices configured to store specific location information indicating specific locations where the remote support may be required; And One or more processors, The one or more processors are configured to: Determine whether there is an abnormality in the remote support system, where the remote support system is configured to provide the remote support to the autonomous vehicle; In the case where the abnormality of the remote support system is detected, based on the specific location information, set any of the specific locations on the target route from the current position of the autonomous vehicle to the destination as the boundary position; Set a target avoidance position such that the target avoidance position is included in the target route from the current position to the boundary position; and Control the autonomous vehicle to stop at the target avoidance position, The one or more processors are further configured to: Based on the specific location information, obtain a first specific location on the target route that is closest to the current position; Determine whether the first specific location satisfies an allowable condition, where the allowable condition includes at least one of the following (i) and (ii): (i) the distance between the current position and the first specific location is equal to or greater than a distance threshold, (ii) the vehicle control amount required to stop the autonomous vehicle in front of the first specific location is equal to or less than a control amount threshold; In the case where the first specific location does not satisfy the allowable condition, based on the specific location information, obtain a second specific location on the target route that is different from the first specific location; and Set the second specific location as the boundary position.

2. The autonomous driving system according to claim 1, wherein The one or more processors are further configured to: Based on the specific location information, obtain a first specific location on the target route that is closest to the current position; and Set the first specific location as the boundary position.

3. The autonomous driving system according to claim 1, wherein The one or more processors are further configured to: Based on the specific location information, obtain a first specific location on the target route that is closest to the current position; and In the case where the abnormality of the remote support system is a malfunction, set the first specific location as the boundary position.

4. The autonomous driving system according to claim 1, wherein The one or more processors are further configured to: Based on the specific location information, obtain a first specific location on the target route that is closest to the current position; When the abnormality of the remote support system is a performance degradation, it is determined whether the first specific position satisfies the allowable condition, where the allowable condition includes at least one of the following (i) and (ii): (i) the distance between the current position and the first specific position is equal to or greater than a distance threshold; (ii) the vehicle control amount required to stop the autonomous driving vehicle in front of the first specific position is equal to or less than a control amount threshold. When the first specific position does not satisfy the allowable condition, a second specific position different from the first specific position on the target route is obtained based on the specific position information. And The second specific position is set as the boundary position.

5. An autonomous driving control method, executed by one or more processors, for controlling an autonomous driving vehicle that is an object of remote support, the autonomous driving control method being characterized by including: Determining whether there is an abnormality in a remote support system that provides the remote support to the autonomous driving vehicle. When the abnormality of the remote support system is detected, based on specific position information, an arbitrary specific position on the target route from the current position of the autonomous driving vehicle to the destination is set as a boundary position, where the specific position information indicates the specific position where the remote support may be required. Setting a target avoidance position such that the target avoidance position is included in the target route from the current position to the boundary position; and Controlling the autonomous driving vehicle to stop at the target avoidance position. The autonomous driving control method further includes: Obtaining a first specific position on the target route that is closest to the current position based on the specific position information. Determining whether the first specific position satisfies the allowable condition, where the allowable condition includes at least one of the following (i) and (ii): (i) the distance between the current position and the first specific position is equal to or greater than a distance threshold; (ii) the vehicle control amount required to stop the autonomous driving vehicle in front of the first specific position is equal to or less than a control amount threshold. When the first specific position does not satisfy the allowable condition, a second specific position different from the first specific position on the target route is obtained based on the specific position information; and The second specific position is set as the boundary position.

6. A non-transitory recording medium storing instructions that can be executed by a computer and cause the computer to execute the following functions for controlling an autonomous driving vehicle that is an object of remote support: Determine the abnormality of the remote support system, where The remote support system provides the remote support to the autonomous driving vehicle. When the abnormality of the remote support system is detected, based on specific position information, an arbitrary specific position on the target route from the current position of the autonomous driving vehicle to the destination is set as a boundary position, where the specific position information indicates the specific position where the remote support may be required. Set a target avoidance position such that the target avoidance position is included in the target route from the current position to the limit position; Control the autonomous vehicle to stop at the target avoidance position; Obtain a first specific position closest to the current position on the target route based on the specific position information; Determine whether the first specific position satisfies an allowable condition, where the allowable condition includes at least one of the following (i) and (ii): (i) the distance between the current position and the first specific position is equal to or greater than a distance threshold, (ii) the vehicle control amount required to stop the autonomous vehicle in front of the first specific position is equal to or less than a control amount threshold; In the case where the first specific position does not satisfy the allowable condition, obtain a second specific position different from the first specific position on the target route based on the specific position information; and Set the second specific position as the limit position.

Images