Vehicle control system, vehicle control method, and computer-readable recording medium

Abstract

The invention relates to a vehicle control system, a vehicle control method, and a computer-readable recording medium. The present invention suppresses the behavior of a vehicle from becoming unstable when switching from a main system to a redundant system when performing travel control of the vehicle using a computing device of the main system and the redundant system. When an abnormality in a first target trajectory generation function of an arithmetic device for generating a first target trajectory is detected, a fail-switch process is performed for switching a base target trajectory used for generating a final target trajectory from the first target trajectory to a second target trajectory. In the fail-over process, a first target trajectory generated before an abnormality of the generation function is detected is updated on the basis of an estimated amount of movement of the vehicle during the fail-over process. Then, a position where at least the target position coincides among the second target trajectory generated after the abnormality has been detected and the updated first target trajectory is set as a switching position of the base target trajectory.

Description

Technical Field

[0001] The present disclosure relates to a system, method, and computer-readable recording medium for controlling a vehicle performing autonomous driving. Background Art

[0002] Japanese Patent Application Laid-Open No. 2019-171970 discloses a system for controlling vehicle driving, wherein the vehicle performs autonomous driving. The system includes computing devices of a main system and a redundant system. The computing device of the main system calculates a target trajectory for driving control. The computing device of the redundant system replaces the computing device of the main system in calculating the target trajectory if an abnormality occurs in the computing device of the main system.

[0003] Japanese Patent Application Laid-Open No. 2021-127039 discloses a system for autonomous vehicle operation. The system includes a primary system and a redundant system driving control unit. The primary system driving control unit performs autonomous vehicle operation. Upon detecting a functional degradation in the primary system driving control unit, the primary system driving control unit transmits an autonomous driving handover instruction to the redundant system driving control unit. The redundant system driving control unit then controls vehicle operation based on the information contained in the handover instruction.

[0004] Japanese Patent Application Laid-Open No. 2021-075186 discloses a system for controlling the driving of a vehicle, wherein the vehicle performs autonomous driving. The system includes multiple computing devices. Each of the computing devices generates a target trajectory for the vehicle during driving control. If a defect occurs in any of the multiple computing devices, the system performs driving control based on the target trajectory generated by computing devices other than the defective computing device.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Laid-Open No. 2019-171970

[0008] Patent Document 2: Japanese Patent Application Laid-Open No. 2021-127039

[0009] Patent Document 3: Japanese Patent Application Laid-Open No. 2021-075186

[0010] The system described in Japanese Patent Application Laid-Open No. 2021-075186 is classified as a hot standby system, in which the computing devices of the primary and redundant systems are always in operation. In contrast, the systems described in Japanese Patent Application Laid-Open Nos. 2019-171970 and 2021-127039 are classified as warm standby systems, in which the computing devices of the primary system are in operation while the computing devices of the redundant system are on standby.

[0011] In hot standby mode, problems caused by insufficient resources such as the CPU (Central Processing Unit) and memory can occur simultaneously in both the primary and redundant systems. Furthermore, in hot standby mode, target trajectories are generated independently by the primary and redundant systems, resulting in completely different target trajectories. Consequently, a sudden switch from the primary to the redundant system can cause the vehicle to accelerate or decelerate suddenly, or even swerve suddenly.

[0012] In this regard, warm standby mode minimizes the resources allocated to the redundant system, allowing it to be placed on standby. By providing handover information from the primary system to the redundant system during the switchover, vehicle behavior can be stabilized during the switchover. However, compared to hot standby mode, warm standby mode has the disadvantage of requiring more time to switch from the primary system to the redundant system. This can lead to a discrepancy between the vehicle's state at the time of handover and the current state, resulting in unstable vehicle behavior. Summary of the Invention

[0013] An object of the present disclosure is to provide a technique for suppressing unstable vehicle behavior when switching from a primary system to a redundant system when driving control of a vehicle is performed using computing devices of a primary system and a redundant system.

[0014] A first aspect of the present disclosure is a system for controlling a vehicle that performs autonomous driving, and the system has the following features.

[0015] The system includes a first computing device, a second computing device, a management device, and a control device.

[0016] The first computing device is configured to generate a first target trajectory for the autonomous driving. The second computing device is configured to generate a second target trajectory for the autonomous driving. The management device manages the first computing device and the second computing device. The management device is configured to generate a final target trajectory for the autonomous driving. The control device is configured to perform autonomous driving control of the vehicle based on the final target trajectory.

[0017] The first target trajectory, the second target trajectory, and the final target trajectory are respectively composed of a set of a target position and a target speed of the vehicle.

[0018] The management device is configured to: generate the final target trajectory using the first target trajectory when a function for generating the first target trajectory in the first computing device is normal; and perform failover processing to switch a base target trajectory used for generating the final target trajectory from the first target trajectory to the second target trajectory when an abnormality in the generation function is detected.

[0019] The management device is configured to: in the fault switching process, based on the state information of the vehicle before the abnormality is detected and the last information of the first target trajectory generated before the abnormality is detected, estimate the movement amount of the vehicle during the fault switching process when the vehicle moves according to the first target trajectory constituting the last information; update the first target trajectory constituting the last information based on the estimated movement amount of the vehicle, thereby calculating the updated target position and updated target speed of the vehicle; compare a first set with a second set, wherein the first set is a set representing the updated target position and the updated target speed, and the second set is a set representing the target position and the target speed constituting the second target trajectory generated after the abnormality is detected; and set the position where at least the target position information in the first set and the second set is consistent as the switching position of the basic target trajectory.

[0020] A second aspect of the present disclosure is a method of controlling a vehicle performing automated driving, which has the following features.

[0021] The method includes: generating a first target trajectory in the autonomous driving; generating a second target trajectory in the autonomous driving; generating a final target trajectory in the autonomous driving; and performing autonomous driving control of the vehicle based on the final target trajectory.

[0022] The first target trajectory, the second target trajectory, and the final target trajectory are respectively composed of a set of a target position and a target speed of the vehicle.

[0023] When the function of generating the first target trajectory of the computing device that generates the first target trajectory is normal, the final target trajectory is generated using the first target trajectory.

[0024] When an abnormality in the generation function is detected, a failover process is performed to switch a base target trajectory used for generation of the final target trajectory from the first target trajectory to the second target trajectory.

[0025] The fault switching process includes: estimating the movement amount of the vehicle during the fault switching process when the vehicle moves according to the first target trajectory constituting the last information based on the state information of the vehicle before the abnormality is detected and the last information of the first target trajectory generated before the abnormality is detected; updating the first target trajectory constituting the last information based on the estimated movement amount of the vehicle, thereby calculating the updated target position and updated target speed of the vehicle; comparing a first set with a second set, wherein the first set is a set representing the updated target position and the updated target speed, and the second set is a set representing the target position and the target speed constituting the second target trajectory generated after the abnormality is detected; and setting the position where at least the target position information in the first set and the second set is consistent as the switching position of the basic target trajectory.

[0026] A third aspect of the present disclosure is a computer-readable recording medium storing a program for controlling a vehicle that performs automated driving, and has the following features.

[0027] The program is configured to cause a computer to execute: generating a first target trajectory in the autonomous driving; generating a second target trajectory in the autonomous driving; generating a final target trajectory in the autonomous driving; and performing autonomous driving control of the vehicle based on the final target trajectory.

[0028] The first target trajectory, the second target trajectory, and the final target trajectory are respectively composed of a set of a target position and a target speed of the vehicle.

[0029] When the function of generating the first target trajectory of the computing device that generates the first target trajectory is normal, the final target trajectory is generated using the first target trajectory.

[0030] When an abnormality in the generation function is detected, a failover process is performed to switch a base target trajectory used for generation of the final target trajectory from the first target trajectory to the second target trajectory.

[0031] The fault switching process includes: estimating the movement amount of the vehicle during the fault switching process when the vehicle moves according to the first target trajectory constituting the last information based on the state information of the vehicle before the abnormality is detected and the last information of the first target trajectory generated before the abnormality is detected; updating the first target trajectory constituting the last information based on the estimated movement amount of the vehicle, thereby calculating the updated target position and updated target speed of the vehicle; comparing a first set with a second set, wherein the first set is a set representing the updated target position and the updated target speed, and the second set is a set representing the target position and the target speed constituting the second target trajectory generated after the abnormality is detected; and setting the position where at least the target position information in the first set and the second set is consistent as the switching position of the basic target trajectory.

[0032] Effects of the Invention

[0033] According to the present disclosure, in the event that an abnormality is detected in the generation function of the first target trajectory possessed by the computing device that generates the first target trajectory, a fault switching process is performed to switch the basic target trajectory used for generating the final target trajectory from the first target trajectory to the second target trajectory. In the fault switching process, the first target trajectory generated before the abnormality in the generation function is detected is updated based on the estimated movement amount of the vehicle during the fault switching process. Then, the position where at least the target position of the second target trajectory generated after the abnormality is detected and the updated first target trajectory are consistent is set as the switching position of the basic target trajectory. Therefore, according to the present disclosure, it is possible to suppress the occurrence of sudden acceleration or deceleration of the vehicle or sudden steering of the vehicle after completion of the fault switching process. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a conceptual diagram illustrating an outline of an embodiment of the present disclosure.

[0035] Figure 2 This is a diagram for explaining the focus of the embodiment.

[0036] Figure 3 This is a diagram explaining the processing performed in the failover processing according to the embodiment.

[0037] Figure 4 This is a block diagram showing a configuration example of a system according to an embodiment.

[0038] Figure 5 This is a flowchart showing the flow of information processing particularly related to the embodiment.

[0039] Figure 6This is a flowchart showing the flow of information processing particularly related to the embodiment.

[0040] Figure 7 This is a flowchart showing the flow of information processing particularly related to the embodiment.

[0041] Description of Reference Numerals

[0042] 10: Information acquisition device; 20: First computing device; 21, 31, 41, 51: Processor; 22, 32, 42, 52: Storage device; 30: Second computing device; 40: Management device; 50: Control device; 60: Travel device; 100: System; CON: Requested control amount; ENV: Driving environment information; ST1, ST2: Action state information; TR: Target trajectory; TR1: First target trajectory; TR2: Second target trajectory; TRB: Basic target trajectory; TRF: Final target trajectory; VH: Vehicle. DETAILED DESCRIPTION

[0043] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0044] 1. Summary

[0045] Figure 1 This is a conceptual diagram for explaining the outline of the system according to the embodiment. Figure 1 The illustrated system 100 is a system for controlling a vehicle VH. Typically, the system 100 is installed in the vehicle VH. Alternatively, some of the functions of the system 100 may be performed by an external device, or the vehicle VH may be controlled by a remote device. In other words, the functions of the system 100 may be distributed between the vehicle VH and an external device.

[0046] The vehicle VH is a vehicle that performs autonomous driving. Autonomous driving in this context assumes that the operator of the vehicle VH (e.g., the driver) does not necessarily have to be fully focused on driving. Autonomous driving assumes so-called Level 3 or higher autonomous driving.

[0047] Autonomous driving of vehicle VH is achieved by system 100. System 100 controls the steering, acceleration, and deceleration of vehicle VH as vehicle control for autonomous driving (hereinafter also referred to as "autonomous driving control"). In autonomous driving control, the driving devices (steering device, drive device, and braking device) of vehicle VH are controlled so that vehicle VH follows a target trajectory TR.

[0048] The target trajectory TR includes a set of target positions [Xi, Yi] of the vehicle VH in the lane in which the vehicle VH is traveling and a set of target speeds [VXi, VYi] for each target position [Xi, Yi]. Figure 1 In the example shown, the X direction is the direction in front of the vehicle VH, and the Y direction is the direction of the plane perpendicular to the X direction. Figure 1 Example shown.

[0049] To make the vehicle VH follow the target trajectory TR, the automatic driving control calculates the deviation (e.g., lateral deviation, yaw angle deviation, and speed deviation) between the vehicle VH and the target trajectory TR, and then controls the driving device of the vehicle VH to reduce the deviation.

[0050] In the embodiments, a case where a target trajectory TR is generated by a system in warm standby mode is considered. In this case, the target trajectory TR is generated by the computing device of the primary system. If an abnormality is detected in the computing device of the primary system, the generation of the target trajectory TR by the computing device of the redundant system is initiated. Automatic driving control is performed based on the target trajectory TR generated by the computing device of the primary system or the redundant system.

[0051] Hereinafter, the computing device of the primary system is also referred to as the "first computing device," and the computing device of the redundant system is also referred to as the "second computing device." The target trajectory TR used for autonomous driving control is also referred to as the "final target trajectory TRF." The target trajectory TR generated by the first computing device is referred to as the "first target trajectory TR1," and the target trajectory TR generated by the second computing device is referred to as the "second target trajectory TR2."

[0052] The first target trajectory TR1, the second target trajectory TR2, and the final target trajectory TRF each include a set of target positions [Xi, Yi] and a set of target velocities [VXi, VYi] for each target position [Xi, Yi]. Hereinafter, the target positions and target velocities included in the first target trajectory TR1 are represented by [X1i, Y1i] and [VX1i, VY1i], respectively. The target positions and target velocities included in the second target trajectory TR2 are represented by [X2i, Y2i] and [VX2i, VY2i], respectively.

[0053] The target position [Xi, Yi] and target speed [VXi, VYi] information included in the final target trajectory TRF generated using the first target trajectory TR1 is consistent with the target position [X1i, Y1i] and target speed [VX1i, VY1i]. The target position [Xi, Yi] and target speed [VXi, VYi] information included in the final target trajectory TRF generated using the second target trajectory TR2 is consistent with the target position [X2i, Y2i] and target speed [VX2i, VY2i].

[0054] Figure 2This is a diagram for explaining the focus of the embodiment. Figure 2 This is also a diagram for explaining the relationship among the final target trajectory TRF, the first target trajectory TR1, and the second target trajectory TR2. Figure 2 Time T1 shown is when an abnormality is detected in the function generating the first target trajectory TR1 in the first computing device. Examples of abnormalities in the generation function include resource exhaustion, such as CPU and memory, and OS (Operating System) processing delays. Therefore, depending on the nature of the abnormality, the first target trajectory TR1 may continue to be generated in the first computing device even after time T1.

[0055] The generation of the second target trajectory TR2 starts at time T1. Also, from time T1, the fail-over process (hereinafter also referred to as "F / O process") starts. The F / O process is a process for switching the target trajectory TR used for generating the final target trajectory TRF (hereinafter also referred to as "base target trajectory TRB") from the first target trajectory TR1 to the second target trajectory TR2. Figure 2 In the example shown, the F / O process is completed at time T2. That is, the F / O process is performed from time T1 to time T2. It should be noted that the interval between time T1 and time T2 depends on the system configuration, and in the embodiment, a fixed time is assumed.

[0056] Here, before the F / O process starts, that is, before time T1, the final target trajectory TRF is generated using the first target trajectory TR1. After the F / O process is completed, that is, after time T2, the final target trajectory TRF is generated using the second target trajectory TR2.

[0057] During the F / O process, that is, between time T1 and time T2, the final target trajectory TRF is not generated. As described above, the second target trajectory TR2 is generated after time T1. The initial generation of the second target trajectory TR2 is performed using the final information of the first target trajectory TR1 generated before time T1 (for example, time T3). This initial generation itself is performed before time T2. However, during the F / O process, the final target trajectory TRF is not generated using the initial second target trajectory TR2. It should be noted that time T3 is time T1 or before time T1.

[0058] Automatic driving control during F / O processing is performed based on the final target trajectory TRF generated using the final information from the first target trajectory TR1. A problem arises here: vehicle VH is moving during F / O processing. Therefore, even if it is assumed that F / O processing is completed in the shortest possible time, the information used to generate the final target trajectory TRF accompanying the completion of F / O processing may deviate from the information reflecting the state of vehicle VH during F / O processing. In particular, in a system operating in warm standby mode, F / O processing sometimes takes time, which can increase this deviation. Furthermore, this deviation increases when vehicle VH's speed or steering angle are high.

[0059] Therefore, in the embodiment, the following processing is performed in the F / O processing. Figure 3 This is a diagram for explaining the processing performed in the F / O processing of the embodiment. Figure 3 The first target trajectory TR1 (T3) generated at time T3 is depicted on the left side of Figure 3 The second target trajectory TR2(T3) generated based on the final information of the first target trajectory TR1 generated at time T3 is depicted to the right. The target position [X1i, Y1i] and target speed [V1Xi, VY1i] information included in the first target trajectory TR1(T3) match the target position [X2i, Y2i] and target speed [VX2i, VY2i] information included in the second target trajectory TR2(T3).

[0060] During the F / O process, the amount of movement of the vehicle VH during the F / O process (in each of the X and Y directions) is estimated based on the history of vehicle VH state information (e.g., vehicle speed, acceleration, and steering angle) prior to time T1. This estimation is based on the assumption that the vehicle VH moves along the final target trajectory TRF generated using the first target trajectory TR1 (T3) during the F / O process. Once the estimated amount of movement of the vehicle VH is calculated, the target position [X1i, Y1i] that the vehicle VH will reach or pass during the F / O process is determined.

[0061] Furthermore, in the F / O process, the first target trajectory TR1 (T3) is updated (corrected) using the determined target position [X1i, Y1i]. The first target trajectory TR1 (T3) is updated, for example, by deleting the information of the determined target position [X1i, Y1i] from the information of the first target trajectory TR1 (T3). Figure 3The first target trajectory TR1*(T3) in the center represents the updated first target trajectory TR1(T3). The target position and target speed (updated target position and updated target speed) included in the first target trajectory TR1*(T3) are represented by [X1*i, Y1*i] and [VX1*i, VY1*i].

[0062] During the F / O process, the set of target positions [X1*i, Y1*i] and target velocities [VX1*i, VY1*i] is compared with the set of target positions [X2i, Y2i] and target velocities [VX2i, VY2i]. The position [X, Y] where at least the target position information matches is set as the switching position from the first target trajectory TR1 to the second target trajectory TR2.

[0063] The target position information consistency mentioned here includes not only the case where the target position [X1*i, Y1*i] is completely consistent with the target position [X2i, Y2i], but also the case where there is an error (a few percent) that can be considered as complete consistency. In addition, the target position information consistency at least may also refer to the case where the target position and target speed information are consistent.

[0064] By setting this switching position, the difference between the target position [X1i, Y1i] before the base target trajectory TRB is switched and the target position [X2i, Y2i] after the base target trajectory TRB is switched can be reduced to a value below the error that allows for complete agreement. This can prevent sudden acceleration or deceleration of the vehicle VH, or abrupt steering of the vehicle VH, after the F / O process is completed. An example system configuration for implementing this function is described below.

[0065] 2. Example of composition

[0066] Figure 4 This is a block diagram illustrating an example configuration of a system 100 according to an embodiment. System 100 includes an information acquisition device 10, a first computing device 20, a second computing device 30, a management device 40, a control device 50, and a driving device 60. These devices are connected via, for example, an in-vehicle network (e.g., a CAN (Controller Area Network)).

[0067] The information acquisition device 10 acquires driving environment information ENV. The driving environment information ENV is information indicating the driving environment of the vehicle VH and is information required for autonomous driving of the vehicle VH. Examples of the driving environment information ENV include map information, position information, status information, surrounding information, and distribution information.

[0068] Map information includes, for example, information on the location and shape of roads. Map information is obtained from an onboard map database or an external server. Position information indicates the position and orientation of the vehicle VH. Position information is obtained, for example, from a GPS (Global Positioning System) device. State information indicates the state of the vehicle VH. State information is obtained from state sensors such as a vehicle speed sensor, yaw rate sensor, acceleration sensor, and steering angle sensor.

[0069] Surrounding information is information about the conditions surrounding the vehicle VH. This information is acquired from recognition sensors such as cameras, LIDAR (Laser Imaging Detection and Ranging), and radar. This information includes information about objects surrounding the vehicle VH (e.g., vehicles, pedestrians, roadside objects, obstacles, white lines, etc.). It also includes information about the objects' relative position and speed relative to the vehicle VH. Distribution information is acquired from external devices. Distribution information is provided by the infrastructure surrounding the vehicle VH and by vehicles surrounding the vehicle VH.

[0070] The first computing device 20 is a device that performs various information processing operations. For example, the first computing device 20 is a microcomputer. The first computing device 20 is also called an ECU (Electronic Control Unit). More specifically, the first computing device 20 includes at least one processor 21 and at least one storage device 22.

[0071] The storage device 22 stores various information. For example, the storage device 22 stores driving environment information ENV acquired by the information acquisition device 10. Examples of the storage device 22 include volatile memory, nonvolatile memory, and a hard disk drive (HDD). The processor 21 executes a computer program. This computer program is stored in the storage device 22 or recorded on a computer-readable recording medium. The functions of the first computing device 20 are realized by the processor 21 executing the computer program.

[0072] The first computing device 20 includes functions for generating a first target trajectory TR1. The first target trajectory TR1 is generated based on driving environment information ENV. For example, the first computing device 20 generates the first target trajectory TR1 for maintaining the vehicle's current lane. As another example, the first computing device 20 generates the first target trajectory TR1 for lane changes. Furthermore, as another example, the first computing device 20 generates the first target trajectory TR1 for avoiding collisions between the vehicle VH and surrounding objects.

[0073] The first computing device 20 (processor 21) outputs the first target trajectory TR1 to the management device 40. Every time the first target trajectory TR1 is generated, the first computing device 20 outputs the latest first target trajectory TR1 to the management device 40.

[0074] The second computing device 30 is a device that performs various information processing. The basic configuration of the second computing device 30 is the same as that of the first computing device 20. Specifically, the second computing device 30 includes at least one processor 31 and at least one storage device 32.

[0075] The storage device 32 stores various information. For example, the storage device 32 stores driving environment information ENV. The processor 31 executes a computer program. The computer program is stored in the storage device 32 or recorded on a computer-readable recording medium. The functions of the second computing device 30 are realized by the processor 31 executing the computer program.

[0076] The second computing device 30 includes a function for generating a second target trajectory TR2. The second target trajectory TR2 is generated based on the driving environment information ENV. The example of the second target trajectory TR2 is the same as the example described for the first target trajectory TR1. It should be noted that when performing F / O processing, the initial generation of the second target trajectory TR is based on the latest information from the first target trajectory TR1. Furthermore, upon receiving feedback information (hereinafter also referred to as "F / B information"), the latest second target trajectory TR2 is updated based on this F / B information.

[0077] The second computing device 30 (processor 31) outputs the second target trajectory TR2 to the management device 40. Each time the second target trajectory TR2 is generated, the second computing device 30 outputs the latest second target trajectory TR2 to the management device 40. If the latest second target trajectory TR2 is updated based on the F / B information, the second computing device 30 outputs the updated second target trajectory TR2 as the latest second target trajectory TR2 to the management device 40.

[0078] The management device 40 is a device that performs various information processing. The basic configuration of the management device 40 is the same as that of the first computing device 20. Specifically, the management device 40 includes at least one processor 41 and at least one storage device 42.

[0079] The storage device 42 stores various information. For example, in addition to the driving environment information ENV, the storage device 42 also stores operating status information ST1 and operating status information ST2. Operating status information ST1 indicates the operating status of the first computing device 20. Operating status information ST2 indicates the operating status of the second computing device 30. The operating status includes information such as CPU usage, memory usage, and network load. The management device 40 manages the first computing device 20 based on the operating status information ST1 and manages the second computing device 30 based on the operating status information ST2.

[0080] The various information stored in the storage device 42 also includes the first target trajectory TR1 and the second target trajectory TR2. The processor 41 executes a computer program. The computer program is stored in the storage device 42 or recorded on a computer-readable recording medium. The functions of the management device 40 are realized by the processor 41 executing the computer program.

[0081] The management device 40 functions include generating a final target trajectory TRF. While the first computing device 20 (particularly, generating the first target trajectory TR1) is functioning normally, the first target trajectory TR1 is used as the target trajectory for generating the final target trajectory TRF. If an abnormality is detected in the first computing device 20, the second target trajectory TR2 is used to generate the final target trajectory TRF.

[0082] The management device 40 outputs the final target trajectory TRF to the control device 50. Every time the final target trajectory TRF is generated, the management device 40 outputs the latest final target trajectory TRF to the control device 50.

[0083] If an abnormality is detected in the function of the first computing device 20, the management device 40 (processor 41) performs F / O processing. This F / O processing may be performed after confirming that the function of the second computing device 30 is normal. Alternatively, if an abnormality is detected in the function of the second computing device 30, the management device 40 may perform additional F / O processing. This additional F / O processing switches the target trajectory TR used to generate the final target trajectory TRF from the second target trajectory TR2 to another target trajectory TR. As an example of another target trajectory TR, the first target trajectory TR1 is shown when the function of the first computing device 20 has been restored.

[0084] In the F / O process, when an abnormality is detected in the function of the first computing device 20, the management device 40 detects the abnormality based on the time ( Figure 3The management device 40 calculates the estimated movement amount of the vehicle VH during the F / O process based on the history of the vehicle VH's state information before time T1 (time T1). Furthermore, the management device 40 generates a first target trajectory TR1(T3) based on the latest information of the first target trajectory TR1 generated before time T1 (time T3). The management device 40 then determines the target position [X1i, Y1i] that the vehicle VH will reach or pass during the F / O process based on the estimated movement amount of the vehicle VH and the first target trajectory TR1(T3).

[0085] In the F / O process, the management device 40 also updates (corrects) the first target trajectory TR1 (T3) using the determined target position [X1i, Y1i]. Then, the management device 40 updates (corrects) the first target trajectory TR1 (T3) (that is, Figure 3 The first target trajectory TR1*(T3) described in is compared with the second target trajectory TR2. The second target trajectory TR2 to be compared is generated based on the last information of the first target trajectory TR1 generated at time T3 and is sent from the second computing device 30 to the management device 40.

[0086] When comparing the updated first target trajectory TR1 (T3) with the second target trajectory TR2, the management device 40 identifies a position [X, Y] where at least the target position information matches. After the vehicle VH reaches the identified position [X, Y], the management device 40 terminates the F / O processing. In parallel with the comparison of the updated first target trajectory TR1 (T3) with the second target trajectory TR2, the management device 40 transmits the information on the updated first target trajectory TR1 (T3) to the second computing device 30 as F / B information.

[0087] When the management device 40 performs additional F / O processing, the basic processing is the same as the above-described F / O processing. The additional F / O processing is described by replacing "first target trajectory TR1" with "second target trajectory TR2," "first target trajectory TR1(T3)" with "second target trajectory TR2(T3)," and "second target trajectory TR2" with "other target trajectory" in the above description of the F / O processing.

[0088] The control device 50 is a device that performs various information processing. The basic configuration of the control device 50 is the same as that of the first computing device 20. Specifically, the control device 50 includes at least one processor 51 and at least one storage device 52.

[0089] The storage device 52 stores various information. For example, the final target trajectory TRF is stored in the storage device 52 . The processor 51 executes a computer program. The computer program is stored in the storage device 52 or recorded on a computer-readable recording medium. The functions of the control device 50 are realized by the processor 51 executing the computer program.

[0090] The functions of the control device 50 include an automatic driving control function. The control device 50 performs automatic driving control by controlling the operation of the travel device 60. Specifically, the control device 50 controls the steering (steering) of the vehicle VH by controlling the operation of the steering device 61. Furthermore, the control device 50 controls the acceleration of the vehicle VH by controlling the operation of the drive device 62. Furthermore, the control device 50 controls the deceleration of the vehicle VH by controlling the operation of the brake device 63.

[0091] In particular, the control device 50 performs autonomous driving control so that the vehicle VH follows the final target trajectory TRF. In this case, the control device 50 calculates the deviation between the vehicle VH and the final target trajectory TRF based on the final target trajectory TRF and the position and state information of the vehicle VH. Examples of deviations include lateral deviation (Y-direction deviation), yaw angle deviation (azimuth deviation), and speed deviation. The control device 50 then performs autonomous driving control so that the deviation between the vehicle VH and the final target trajectory TRF is minimized.

[0092] The control device 50 calculates a control variable for controlling the travel device 60, namely, a control variable for at least one of steering, acceleration, and deceleration. Hereinafter, the control variable requested to cause the vehicle VH to follow the final target trajectory TRF, that is, the control variable requested to reduce the deviation between the vehicle VH and the final target trajectory TRF, is referred to as the "requested control variable CON." Examples of the requested control variable CON include a target steering angle, a target yaw rate, a target speed, a target acceleration, a target deceleration, a target torque, and a target current. The control device 50 controls the operation of the travel device 60 according to the requested control variable CON. In other words, the control device 50 controls at least one of steering, acceleration, and deceleration.

[0093] The travel device 60 includes a steering device 61, a drive device 62, and a brake device 63. The steering device 61 steers the wheels of the vehicle VH. For example, the steering device 61 includes an electric power steering (EPS) device. The drive device 62 is a power source that generates driving force. Examples of the drive device 62 include an engine, an electric motor, and an in-wheel motor. The brake device 63 generates braking force.

[0094] 3. Information Processing Examples

[0095] Figure 5 - Figure 7 This is a flowchart showing the flow of processing particularly related to the embodiment. Figure 5 Related to the processing executed by the first arithmetic device 20 (processor 21). Figure 6 Related to the processing executed by the management device 40 (processor 41). Figure 7 Related to the processing executed by the second arithmetic device 30 (processor 31). Figure 5 - Figure 7 The flowcharts shown are repeatedly executed at regular intervals.

[0096] exist Figure 5 In the illustrated routine, information acquisition processing is first performed (step S11). During this information acquisition process, for example, driving environment information ENV is acquired. Furthermore, during this information acquisition process, operational status information ST1 is acquired. This information acquisition process also acquires external information. The external information referred to here includes, for example, input information from computing devices other than the first computing device 20, such as input information from the second computing device 30 and the management device 40.

[0097] Following the processing of step S11, it is determined whether a switch command for the base target trajectory TRB (hereinafter also referred to as a "TR switch command") has been received (step S12). The TR switch command corresponds to the information input from the management device 40 to the first computing device 20, among the external information described above. The TR switch command is output from the management device 40 when it is determined that F / O processing is to be started. In other words, the output of the TR switch command occurs when an abnormality is detected in the function of the first computing device 20.

[0098] If the result of step S12 is negative, a first target trajectory TR1 is generated based on the driving environment information ENV (step S13). An example of the first target trajectory TR1 generated based on the driving environment information ENV is described above. The generated first target trajectory TR1 is then output (step S14). The output destination of the first target trajectory TR1 is the management device 40.

[0099] If the result of step S12 is affirmative, that is, if a TR switching command has been received, failover information (hereinafter also referred to as "F / O information") is output (step S15). The F / O information includes the most recent information about the first target trajectory TR1 at the time of step S12. In other words, the F / O information includes the latest information about the first target trajectory TR1 generated at the time of step S12. The F / O information is output to the second computing device 30 and the management device 40.

[0100] exist Figure 6 In the illustrated routine, information acquisition processing is first performed (step S21). During this information acquisition process, for example, driving environment information ENV is acquired. Furthermore, during this information acquisition process, operational status information ST1 and operational status information ST2 are acquired. This information acquisition process also acquires external information. The external information referred to here includes, for example, input information from computing devices other than the management device 40, such as input information from the first computing device 20, the second computing device 30, and the control device 50.

[0101] Following step S21, a determination is made as to whether an abnormality has been detected in the function of the first computing device 20 (step S22). Step S22 determines whether to initiate F / O processing. Step S22 is performed based on the operational status information ST1 acquired in step S21. If the determination in step S22 is negative, meaning no abnormality has been detected in the function of the first computing device 20, a final target trajectory TRF is generated using the latest first target trajectory TR1 (step S23). After the final target trajectory TRF is generated, it is output to the control device 50 (step S29).

[0102] If the result of the determination in step S22 is affirmative, a TR switching command is output (step S24). The TR switching command is output to the first computing device 20 and the second computing device 30. The TR switching command to the first computing device 20 includes a command to stop the process of generating the first target trajectory TR1 and a command to output the latest information of the first target trajectory TR1 at the current time point to the second computing device 30. The TR switching command to the second computing device 30 includes a command to start the process of generating the second target trajectory TR.

[0103] In another example of the process of step S24, the TR switching instruction to the first computing device 20 may simply instruct the first target trajectory TR1 to be stopped. In this case, the TR switching instruction to the second computing device 30 includes the latest information on the first target trajectory TR1 currently held by the management device 40 and an instruction to start the second target trajectory TR generation process based on this latest information.

[0104] Following step S24, the F / O process is performed (step S25). The details of the F / O process are described above. During the F / O process, the target position [X1i, Y1i] that the vehicle VH will reach or pass during the F / O process is determined, and the first target trajectory TR1 (T3) is updated (corrected) using this target position [X1i, Y1i]. The updated first target trajectory TR1 (T3) is output as F / B information (step S26). The F / B information is output to the second computing device 30.

[0105] Following step S26, it is determined whether the switching of the base target trajectory TRB has completed (step S27). As described above, in the F / O process, the switching position [X, Y] from the first target trajectory TR1 to the second target trajectory TR2 is set. Then, after the vehicle VH reaches this switching position [X, Y], the F / O process ends. The process of step S27 is repeated until the vehicle VH reaches the switching position [X, Y].

[0106] If the result of the determination in step S27 is affirmative, the final target trajectory TRF is generated using the latest second target trajectory TR2 (step S28 ). After the final target trajectory TRF is generated, it is output to the control device 50 (step S29 ).

[0107] exist Figure 7 In the illustrated routine, information acquisition processing is first performed (step S31). During this information acquisition process, for example, driving environment information ENV is acquired. Furthermore, during this information acquisition process, operational status information ST2 is acquired. This information acquisition process also acquires external information. The external information referred to here includes, for example, input information from computing devices other than the second computing device 30, such as input information from the first computing device 20 and the management device 40.

[0108] Following the processing of step S31, it is determined whether a TR switching command has been received (step S32). The TR switching command corresponds to the information input from the management device 40 to the second computing device 30, among the external information described above. If the determination result of step S32 is negative, the processing routine ends. It should be noted that in this case, the second computing device 30 (processor 31) waits for the generation of the second target trajectory TR2.

[0109] If the result of step S32 is affirmative, a second target trajectory TR2 is generated based on the F / O information (step S33). The F / O information is input from the first computing device 20 immediately after receiving the TR switching command, or from the management device 40 simultaneously with the TR switching command. The generated second target trajectory TR2 is then output (step S34). The output destination of the second target trajectory TR2 is the management device 40.

[0110] Following the processing of step S34, it is determined whether the F / B information is received (step S35). As described above, the F / B information is generated in the management device 40 during the F / O processing and output to the second computing device 30 (see Figure 6 The process of step S35 is repeatedly executed until the second computing device 30 receives the F / B information.

[0111] If the result of the determination in step S35 is positive, the latest second target trajectory TR2 is updated based on the F / B information (step S36). The F / B information is the first target trajectory TR1 (T3) updated during the F / O process (that is, Figure 3 When the F / B information is updated, the latest second target trajectory TR2 information stored in the second computing device 30 is replaced with the first target trajectory TR1*(T3). This ensures that the latest second target trajectory TR2 information stored in the second computing device 30 is consistent with the target trajectory information stored in the management device 40.

[0112] Following the processing of step S36, the updated second target trajectory TR2 (that is, the first target trajectory TR1*(T3)) is output (step S37). The output destination of the second target trajectory TR2 is the management device 40. It should be noted that the generation processing of the second target trajectory TR2 after the processing of step S37 is based on the Figure 5 The process of generating the routine described in the following is explained. Figure 5 In the description of FIG, “the first computing device 20 ” is replaced by “the second computing device 30 ”, and “the first target trajectory TR1 ” is replaced by “the second target trajectory TR2 ”.

[0113] 4. Modifications

[0114] In the embodiments, a warm standby system is assumed. However, the present disclosure can also be applied to a hot standby system. In a hot standby system, the second target trajectory TR2 is constantly generated in the second computing device 30. Therefore, the time required for F / O processing is expected to be shorter than in a warm standby system.

[0115] However, in a hot standby system, when an abnormality is detected in the first computing device 20 ( Figure 3 The latest second target trajectory TR2 (at time T1) may not completely match the latest first target trajectory TR1. Therefore, the above-mentioned F / O processing is performed to update the latest first target trajectory TR1(T3), and the updated first target trajectory TR1(T3) (that is, the first target trajectory TR1*(T3)) is compared with the latest second target trajectory TR2.

[0116] Then, the position [X, Y] where at least the target position information matches among these target trajectories is set as the switching position. Thus, even if the latest second target trajectory TR2 does not completely match the latest first target trajectory TR1, it is possible to prevent sudden acceleration or deceleration of the vehicle VH or abrupt steering of the vehicle VH after the F / O processing is completed.

Claims

1. A vehicle control system for controlling an autonomously driven vehicle, wherein the vehicle control system comprises: a first computing device for generating a first target trajectory in the autonomous driving; a second computing device for generating a second target trajectory in the autonomous driving; a management device for managing the first computing device and the second computing device to generate a final target trajectory in the autonomous driving; as well as A control device performs automatic driving control of the vehicle based on the final target trajectory, The first target trajectory, the second target trajectory, and the final target trajectory are respectively composed of a set of a target position and a target speed of the vehicle, The management device is configured to: When the generation function of the first target trajectory in the first computing device is normal, using the first target trajectory to generate the final target trajectory; as well as When an abnormality in the generation function is detected, a failover process is performed to switch the basic target trajectory used for generating the final target trajectory from the first target trajectory to the second target trajectory. The management device is configured to: in the failover process, estimating, based on the state information of the vehicle before the abnormality is detected and the last information of the first target trajectory generated before the abnormality is detected, an amount of movement of the vehicle during the failover process if the vehicle moves according to the first target trajectory constituting the last information; updating the first target trajectory constituting the final information based on the estimated movement amount of the vehicle, thereby calculating an updated target position and an updated target speed of the vehicle; comparing a first set with a second set, wherein the first set is a set representing the updated target position and the updated target speed, and the second set is a set representing the target position and the target speed constituting the second target trajectory generated after the abnormality is detected; as well as A position where at least target position information in the first set and the second set is consistent is set as a switching position of the basic target trajectory.

2. The vehicle control system according to claim 1, characterized in that: The management device is configured to: in the failover process, A position where both target position and target speed information in the first set and the second set match is set as a switching position of the basic target trajectory.

3. The vehicle control system according to claim 1 or 2, characterized in that: The second computing device is configured to: When the generation function of the first target trajectory is normal, the generation of the second target trajectory is put on hold; initiating generation of the second target trajectory after detection of the anomaly; as well as The second target trajectory is initially generated after the abnormality is detected based on the latest information.

4. The vehicle control system according to claim 1 or 2, characterized in that: The management device is configured to calculate the updated target position and the updated target speed of the vehicle only during the failover process.

5. The vehicle control system according to claim 1 or 2, characterized in that: The management device sends information of the updated target position and updated target speed calculated during the failover process to the second computing device. Upon receiving the information of the updated target position and the updated target speed calculated during the fault switching process, the second computing device updates the second target trajectory generated before receiving the information of the updated target position and the updated target speed based on the received information of the updated target position and the updated target speed.

6. A vehicle control method for controlling an autonomously driving vehicle, the vehicle control method comprising: generating a first target trajectory in the autonomous driving; generating a second target trajectory in the autonomous driving; generating a final target trajectory in the autonomous driving; as well as performing autonomous driving control of the vehicle based on the final target trajectory, The first target trajectory, the second target trajectory, and the final target trajectory are respectively composed of a set of a target position and a target speed of the vehicle, When the generating function of the first target trajectory of the computing device for generating the first target trajectory is normal, the final target trajectory is generated using the first target trajectory. When an abnormality in the generation function is detected, a failover process is performed to switch the basic target trajectory used for generating the final target trajectory from the first target trajectory to the second target trajectory. The failover process includes: estimating, based on the state information of the vehicle before the abnormality is detected and the last information of the first target trajectory generated before the abnormality is detected, an amount of movement of the vehicle during the failover process if the vehicle moves according to the first target trajectory constituting the last information; updating the first target trajectory constituting the final information based on the estimated movement amount of the vehicle, thereby calculating an updated target position and an updated target speed of the vehicle; comparing a first set representing the updated target position and the updated target speed with a second set representing the target position and the target speed constituting the second target trajectory generated after the abnormality is detected; and A position where at least target position information in the first set and the second set is consistent is set as a switching position of the basic target trajectory.

7. A computer-readable recording medium storing a program for controlling an autonomously driven vehicle, wherein the program is configured to cause a computer to execute: generating a first target trajectory in the autonomous driving; generating a second target trajectory in the autonomous driving; generating a final target trajectory in the autonomous driving; as well as performing autonomous driving control of the vehicle based on the final target trajectory, The first target trajectory, the second target trajectory, and the final target trajectory are respectively composed of a set of a target position and a target speed of the vehicle, When the generating function of the first target trajectory of the computing device for generating the first target trajectory is normal, the final target trajectory is generated using the first target trajectory. When an abnormality in the generation function is detected, a failover process is performed to switch the basic target trajectory used for generating the final target trajectory from the first target trajectory to the second target trajectory. The failover process includes: estimating, based on the state information of the vehicle before the abnormality is detected and the last information of the first target trajectory generated before the abnormality is detected, an amount of movement of the vehicle during the failover process if the vehicle moves according to the first target trajectory constituting the last information; updating the first target trajectory constituting the final information based on the estimated movement amount of the vehicle, thereby calculating an updated target position and an updated target speed of the vehicle; comparing a first set representing the updated target position and the updated target speed with a second set representing the target position and the target speed constituting the second target trajectory generated after the abnormality is detected; and A position where at least target position information in the first set and the second set is consistent is set as a switching position of the basic target trajectory.

Images