Autonomous driving system

Abstract

An autonomous driving system includes: a driving plan unit configured to generate a target path in a predetermined coordinate system and a speed plan specifying the transit time at control points on the target path based on a target route, and to reconstruct the speed plan based on the actual speed of the vehicle when an operational intervention is performed during autonomous driving, which is executed by autonomous driving control to make the vehicle travel along the target path according to the speed plan, the operational intervention changing the braking force acting on the vehicle; and a driving control unit configured to perform autonomous driving control.

Description

[0001] This application is a divisional application of Chinese patent application filed on September 26, 2018, with application number 201811123098.4 and invention title "Autonomous Driving System". Technical Field

[0002] This invention relates to autonomous driving systems. Background Technology

[0003] Japanese Patent Application Publication No. 2016-099713 (JP 2016-099713 A) discloses examples of technology related to autonomous driving systems. The autonomous driving system disclosed in this publication generates a driving plan along a preset target route and autonomously controls the driving of the vehicle based on the generated driving plan. Summary of the Invention

[0004] In addition to the autonomous driving system, various safety systems designed to ensure safety during collisions or driving are also installed in the vehicle, including the Pre-Collision System (PCS), Anti-lock Braking System (ABS), Traction Control (TRC), and Vehicle Stability Control (VSC). These safety systems operate the vehicle's brake actuators. When the operation of the autonomous driving system's brake actuators interferes with the operation of the safety system's brake actuators, the safety system's operation takes precedence. Furthermore, even when the autonomous driving system is in operation, the driver can still perform braking actions. The driver's braking actions also take precedence over the autonomous driving system's brake actuator operations. This means that during autonomous driving, the operation of the safety systems and the driver's braking actions constitute operational interventions within the operation of the autonomous driving system's brake actuators.

[0005] The operation of the safety system's brake actuators or the driver's braking action alters the braking force acting on the vehicle, thus changing the vehicle speed. Therefore, if a safety system operation or driver braking action is performed during autonomous driving system operation, the safety system operation or driver braking action will change the vehicle's speed, resulting in a deviation from the driving plan. In the autonomous driving system disclosed in the technology described above, the vehicle's movement is controlled such that, under the assumption that the vehicle will travel at a planned speed, it follows a target path. Therefore, when the speed changes due to safety system operation or driver braking action, there is a possibility that the vehicle's ability to follow the target path may be lost, or the vehicle's behavior may become unstable.

[0006] The present invention provides an autonomous driving system that can stabilize the vehicle's behavior while maintaining its ability to follow the target path, even when an operational intervention that can change the braking force acting on the vehicle is performed during autonomous driving.

[0007] One aspect of the present invention provides an autonomous driving system. The autonomous driving system according to this aspect includes: a driving plan unit configured to generate a target path in a predetermined coordinate system and a speed plan specifying the transit time at control points on the target path based on a target route; and to reconstruct the speed plan based on the actual speed of the vehicle when an operational intervention is performed during autonomous driving, executed by autonomous driving control to cause the vehicle to travel along the target path according to the speed plan, the operational intervention altering the braking force acting on the vehicle; and a driving control unit configured to perform autonomous driving control.

[0008] When an intervention altering the braking force applied to the vehicle is performed during autonomous driving, a difference arises between the planned speed determined by the speed plan and the actual speed. This difference causes a deviation in the control points on the target path to be referenced during autonomous driving control, resulting in a reduction in the vehicle's target path following ability and unstable vehicle behavior. Even when an intervention altering the braking force applied to the vehicle is performed during autonomous driving, the configuration described above allows the speed plan to be reconstructed based on the actual speed. By reconstructing the speed plan in this way to match the planned speed with the actual speed, a reduction in the vehicle's target path following ability is prevented, and unstable vehicle behavior is also prevented, even when such a situation is caused by a difference between the planned and actual speeds.

[0009] In this respect, the driving plan unit can be configured to reconstruct the speed plan based on the vehicle's actual speed and actual acceleration when an operational intervention is performed.

[0010] In this respect, the driving plan unit can be configured to: when an operational intervention is performed, reconstruct the speed plan by matching the planned speed determined by the speed plan with the actual speed and by matching the planned acceleration determined by the speed plan with the actual acceleration.

[0011] The configuration described above also prevents a reduction in the vehicle's target path following ability and unstable vehicle behavior.

[0012] In this respect, the driving plan unit can be configured to match the planned speed determined by the speed plan with the actual speed while the operational intervention is being performed, and to rebuild the speed plan when the operational intervention ends, so that the planned speed gradually increases from the actual speed at the time the operational intervention ends.

[0013] The configuration described above can prevent unstable vehicle behavior caused by sudden changes in acceleration after the termination of operational intervention.

[0014] In this respect, the driving plan unit can be configured to reconstruct a speed plan based on the target acceleration determined by autonomous driving control and the actual speed of the vehicle when the operational intervention is the operation of ABS.

[0015] In this respect, the driving plan unit can be configured to: when the operational intervention is the operation of ABS, reconstruct the speed plan by matching the planned acceleration determined by the speed plan with the target acceleration determined by autonomous driving control and by matching the planned speed determined by the speed plan with the actual speed.

[0016] The configuration described above prevents a reduction in the vehicle's ability to follow the target path and prevents unstable vehicle behavior, while ensuring the maximum possible deceleration under the constraints of road conditions.

[0017] As described above, even when an operational intervention that could alter the braking force acting on the vehicle is performed during autonomous driving, the autonomous driving system according to the present invention can stabilize the vehicle's behavior while maintaining the vehicle's ability to follow the target path. Attached Figure Description

[0018] The features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, in which the same reference numerals denote the same elements, and in the drawings:

[0019] Figure 1 This is a block diagram showing the configuration of the control system of an autonomous vehicle on which an autonomous driving system according to an embodiment of the present invention is mounted.

[0020] Figure 2 This is a diagram illustrating an example of a driving plan generated in an autonomous driving system according to an embodiment of the present invention and the calculation of the amount of operations required for autonomous driving control;

[0021] Figure 3 This is a diagram illustrating the problem of the driving plan of the autonomous driving system in the comparative example;

[0022] Figure 4A It is a diagram showing the target path, the vehicle's current position, and the vehicle's predicted position under normal driving conditions without rebuilding the driving plan;

[0023] Figure 4B This is a diagram illustrating a method for reconstructing a driving plan in an autonomous driving system according to an embodiment of the present invention and its effects;

[0024] Figure 5 This is a diagram illustrating an example of the behavior of a vehicle in an autonomous driving system in a comparative example;

[0025] Figure 6This is a diagram illustrating an example of a method for reconstructing a driving plan in an autonomous driving system according to an embodiment of the present invention, and an example of vehicle behavior;

[0026] Figure 7 This is a diagram illustrating the method of reconstructing the driving plan when TRC is activated; and

[0027] Figure 8 This is a diagram illustrating the method of reconstructing the driving plan when ABS is activated. Detailed Implementation

[0028] Embodiments of the present invention will now be described in detail with reference to the accompanying drawings. When numbers (number of elements, quantity, quantity, range, etc.) are mentioned in the following description of the embodiments, the present invention is not limited to the mentioned numbers unless otherwise expressly stated or clearly indicated. Furthermore, unless otherwise expressly stated or clearly indicated, the structures described in the following embodiments are not necessarily essential to the present invention. 1. Configuration of the control system for an autonomous vehicle

[0029] The autonomous driving system in this embodiment of the invention is a control system for autonomous driving installed on an autonomous driving vehicle. For example, the autonomous driving system is a control system capable of achieving Autonomous Driving Level 3 or higher as defined by the Society of Automotive Engineers (SAE). The autonomous driving vehicle equipped with the autonomous driving system of this embodiment has the following control system, which has… Figure 1 The configuration is shown in the block diagram.

[0030] The autonomous vehicle 100 includes an electronic control unit (ECU) 1 used as a control device. The ECU 1 is a computer having at least one processor and at least one memory. The ECU 1 can perform various functions by having the processor execute programs stored in the memory. The functions implemented include at least the functions of the autonomous driving system 10 and the functions of the safety system 40. Details of the autonomous driving system 10 will be described later.

[0031] Safety system 40 includes: PCS, a safety system for ensuring safety during a collision; and ABS, TRC, and VSC, safety systems for ensuring safety during driving. PCS is a system that predicts a collision with an object ahead and performs vehicle controls to assist in avoiding a collision or to reduce collision damage. ABS is a system that prevents wheel lock-up during braking to ensure steering wheel operability. TRC is a system that reduces wheel slippage during start-up or acceleration to ensure stability. VSC is a system that reduces wheel sideslip to ensure stability. These safety systems 40, located separately from the autonomous driving system 10, at least control the vehicle's braking force.

[0032] GPS unit 2, map information unit 3, and navigation system 4 are connected to the input unit of ECU 1. GPS unit 2 is a unit that acquires location information indicating the current location of the vehicle based on GPS signals. ECU 1 can know the current location of the vehicle based on the location information provided by GPS unit 2. In this specification, unless otherwise stated, "vehicle" refers to the main vehicle on which the autonomous driving system 10 is installed. Map information unit 3 is, for example, a database formed in a storage unit such as an HDD or SSD installed in the vehicle. Map information stored in map information unit 3 includes, for example, road location information, road shape information, location information about intersections and branch points, and road lane information. GPS unit 2 and map information unit 3 are also connected to navigation system 4.

[0033] Navigation system 4 is a device that guides the vehicle to a destination specified by the driver on a map. Navigation system 4 calculates the target route to the destination based on the vehicle's location information measured by GPS unit 2 and map information stored in map information unit 3, and outputs the calculated target route to ECU 1. Note that when ECU 1 is connected to the Internet, the map information can be stored on a server on the Internet, and the target route can be calculated by the server on the Internet.

[0034] Autonomous sensor 5 and vehicle sensor 6 are also connected to the input unit of ECU 1. Autonomous sensor 5 is a sensor that acquires information about the vehicle's surrounding environment and objects. Autonomous sensor 5 includes at least one of, for example, LiDAR (LiDAR: laser imaging detection and ranging), millimeter-wave radar, and a camera. Vehicle sensor 6 is a sensor that acquires information about the vehicle's operating state. Vehicle sensor 6 includes, for example, a speed sensor for measuring the vehicle's speed based on wheel rotation speed, an acceleration sensor for measuring acceleration acting on the vehicle, a yaw rate sensor for measuring the vehicle's angular velocity, and a steering angle sensor for measuring the steering angle.

[0035] In addition, ECU 1 receives signals corresponding to the amount of operation of the driving control tools 50 operated by the driver. These driving control tools include, for example, the brake pedal, accelerator pedal, and steering wheel.

[0036] A drive actuator 7 for driving the vehicle, a steering actuator 8 for steering the vehicle, and a brake actuator 9 for braking the vehicle are connected to the output unit of ECU 1. Drive actuator 7 includes, for example, an engine, an EV system, a hybrid system, and a fuel cell system. Steering actuator 8 includes, for example, a power steering system, a steer-by-wire steering system, and a rear-wheel steering system. Brake actuator 9 includes, for example, a hydraulic brake and a regenerative braking system. These actuators 7, 8, and 9 operate according to operating command values ​​sent from ECU 1.

[0037] 2. Configuration of the autonomous driving system

[0038] Figure 1 This is a detailed block diagram illustrating the functions of ECU 1 when it operates as the autonomous driving system 10. The functions of the autonomous driving system 10 will be described below.

[0039] The autonomous driving system 10 includes a driving plan unit 20 and a driving control unit 30. The driving plan unit 20 is configured to create a driving plan for the vehicle to travel along a target route calculated by the navigation system 4. The driving control unit 30 is configured to perform autonomous driving control according to the driving plan created by the driving plan unit 20. Note that autonomous driving control includes speed control and steering control.

[0040] More specifically, the driving plan unit 20 includes a vehicle status information acquisition unit 21, a vehicle position information acquisition unit 22, a target trajectory information generation unit 23, a target trajectory information update request unit 24, a driver operation determination unit 25, and a safety system operation determination unit 26. First, the vehicle status information acquisition unit 21, the vehicle position information acquisition unit 22, and the target trajectory information generation unit 23 will be described. After describing the driving control unit 30, the target trajectory information update request unit 24, the driver operation determination unit 25, and the safety system operation determination unit 26 will be described.

[0041] Vehicle state information acquisition unit 21 acquires information about vehicle state quantities from vehicle sensor 6. These vehicle state quantities include speed, longitudinal acceleration, yaw rate, and steering angle. Vehicle position information acquisition unit 22 acquires information about vehicle position from GPS unit 2. The vehicle position is represented in an absolute coordinate system, and this position is the location on a map stored in map information unit 3. When autonomous sensor 5 includes a camera, the vehicle's position on the map is corrected by aligning the images captured by the camera with landmarks included in the map information.

[0042] The target trajectory information generation unit 23 acquires the target route calculated by the navigation system 4, the vehicle state variables acquired by the vehicle state variable information acquisition unit 21, and the vehicle position acquired by the vehicle position information acquisition unit 22. Based on the acquired information, the target trajectory information generation unit 23 generates a driving plan for the vehicle to travel along the target route. The generation of the driving plan includes the generation of the target path and the generation of the speed plan.

[0043] A target path is established along the target route; this target path is the route the vehicle travels along for several seconds or tens of seconds. More specifically, the target path is a path formed by connecting the target positions of the vehicle in a predetermined coordinate system; for example, the target path is represented by a set of control points, each with a free X and Y coordinate. The coordinate system representing the target path can be an absolute coordinate system used for map display or a vehicle coordinate system specifically created for the vehicle, in which the X-axis indicates the width direction of the vehicle, and the Y-axis indicates the direction of travel.

[0044] A speed plan is a specification of the transit time for a vehicle at control points on a target path. When determining the transit time at each control point as the vehicle passes through control points sequentially, the transit speed is uniquely determined. This means that specifying the transit time for a vehicle at each control point on the target path is equivalent to specifying the transit speed for the vehicle at each control point on the target path. A speed plan can also be represented as a speed pattern, where a planned speed is set for each control position in a time-dependent manner. Additionally, a speed plan can include an acceleration pattern, where a planned acceleration is set for each control position in a time-dependent manner.

[0045] Figure 2 This is a diagram illustrating an example of a driving plan generated by the target trajectory information generation unit 23. Figure 2 The target path, the vehicle's current position, and the predicted position are shown, all represented in an absolute coordinate system. The vehicle's current position is its position at time t = T0. The predicted position is the vehicle's predicted future location. Figure 2 The diagram shows the predicted positions of the vehicle at times t = T1, T2, and T3. The predicted positions of the vehicle are calculated based on the speed plan, the vehicle's current position, and the vehicle's current state variables.

[0046] Return to Figure 1The description of the target trajectory information generation unit 23 will continue. The target trajectory information generation unit 23 combines the generated target path with the speed plan to form target trajectory information. The target path specifies the location, while the target trajectory information created by combining the speed plan and the target path specifies the location and time. The target trajectory information generation unit 23 provides the generated target trajectory information to the driving control unit 30. Furthermore, the target trajectory information generation unit 23 repeatedly generates target trajectory information based on the latest information at predetermined periodic intervals (e.g., at periodic intervals of several hundred milliseconds). Figure 2 The intervals between the times T0, T1, T2, and T3 shown correspond to the period interval.

[0047] Next, the driving control unit 30 will be described. The driving control unit 30 includes an operation quantity calculation unit 31, a drive control unit 32, a steering control unit 33, and a brake control unit 34. The operation quantity calculation unit 31 calculates the operation quantities for autonomous driving based on target trajectory information provided from the target trajectory information generation unit 23. The operation quantities calculated by the operation quantity calculation unit 31 include operation quantities for speed control and operation quantities for steering control. More specifically, the operation quantity calculation unit 31 calculates the operation quantities of each of the drive actuator 7, brake actuator 9, and steering actuator 8. (Refer to...) Figure 2 Here is an example of the operation quantity calculation method used by the operation quantity calculation unit 31.

[0048] The operation quantity calculation unit 31 determines the operation quantity based on feedforward control and feedback control, enabling the vehicle to travel along the target path according to the speed plan. More specifically, control points on the target path corresponding to a predetermined time prior to the current time are set as reference points. Figure 2 In the example shown, the control point P2 at time T2, two intervals prior to the current time T0, is determined as the reference point for time T0. Once the reference point is determined, a feedforward value for the operational quantity at time T0 is calculated based on parameters corresponding to that reference point. Parameters referenced in calculating the feedforward value include, for example, the curvature of the target path. Furthermore, a feedback correction amount for the operational quantity at time T0 is calculated based on parameters indicating the magnitude of the deviation between the reference point at time T0 and the predicted position at time T2, and parameters indicating the trend. Parameters referenced in calculating the feedback correction amount include, for example, lateral deviation (deviation in the vehicle width direction) and yaw angle deviation. The operational quantity calculation unit 31 calculates the sum of the feedforward value and the feedback correction amount as the operational quantity at time T0.

[0049] Return to Figure 1The description of the driving control unit 30 will continue. The operating quantity of the drive actuator 7, calculated by the operating quantity calculation unit 31, is provided to the drive control unit 32. The operating quantity of the steering actuator 8 is provided to the steering control unit 33. Similarly, the operating quantity of the brake actuator 9 is provided to the brake control unit 34. The drive control unit 32, steering control unit 33, and brake control unit 34 convert the operating quantities provided from the operating quantity calculation unit 31 into operating command values, and send the operating command values ​​to actuators 7, 8, and 9, respectively.

[0050] Note that control units 32, 33, and 34 also accept the operating amounts of actuators 7, 8, and 9 requested by safety system 40. For example, when PCS is activated, brake control unit 34 accepts the operating amount requested by PCS from brake actuator 9 to avoid a collision or reduce collision damage. Similarly, when ABS is activated, brake control unit 34 accepts the operating amount requested by ABS from brake actuator 9 to prevent wheel lock-up.

[0051] In addition, control units 32, 33, and 34 receive the operation amounts of actuators 7, 8, and 9 from the driver's operation request via the driving operation tool 50. For example, when the driver performs a braking operation, the brake control unit 34 receives the operation amount of the brake actuator 9 calculated by switching the brake pedal.

[0052] Each of control units 32, 33, and 34 adds the operation quantity calculated by operation quantity calculation unit 31, the operation quantity requested by safety system 40, and the operation quantity requested by the driver through operation of driving control tool 50, or coordinates among these operation quantities. When coordinating operation quantities, the operation quantities requested by safety system 40 and the operation quantities requested by the driver take precedence over the operation quantities calculated by operation quantity calculation unit 31, i.e., the operation quantities requested by autonomous driving system 10. Therefore, if an operation of safety system 40 or an operation of driving control tool 50 is performed during autonomous driving, the operation of safety system 40 or the operation of driving control tool 50 becomes an operational intervention in the operation of actuators 7, 8, and 9 performed by autonomous driving system 10.

[0053] Note that when an operational intervention caused by the operation of the safety system 40 or by the driver's braking operation is performed during autonomous driving, the vehicle's speed changes, resulting in a deviation from the driving plan. The autonomous driving system 10 in this embodiment has the function to solve this problem. Before describing this function, reference will be made to... Figure 3 This describes the problem of driving planning in an autonomous driving system in a comparative example of applying conventional autonomous driving control.

[0054] like Figure 3As shown, it is assumed that the driver performs a braking operation to brake the vehicle between time T0 and time T1. As a result, the vehicle decelerates through the braking operation, with a deviation between the actual speed of the vehicle and the planned speed determined by the speed plan. The reference point for time T1 in the calculation of the operation amount is the control point on the target path at the previous two intervals, time T3. Note that in the autonomous driving system of the comparative example, the driving plan is generated under the assumption that the vehicle always travels at the planned speed. Therefore, in the autonomous driving system of the comparative example, the control point P2f corresponding to the predicted position (indicated by the dashed line) at time T3 (which is the predicted position when the vehicle has not decelerated) is determined as the reference point for time T1.

[0055] However, when the vehicle decelerates, the correct predicted position of the vehicle at time T3 is indicated by a solid line. In this case, the control point P2t corresponding to this correct predicted position is the correct reference point at time T1. Therefore, when calculating the operation amount using control point P2f as the reference point at time T1, problems arise such as failing to obtain an appropriate feedforward value or the feedback correction becoming too large. This problem may lead to the inability to maintain the vehicle's target path following ability or the vehicle's behavior becoming unstable.

[0056] Return to Figure 1 The functions prepared in the autonomous driving system 10 of this embodiment to solve the above-mentioned problems will be described. The target trajectory information update request unit 24, driver operation determination unit 25, and safety system operation determination unit 26 included in the driving plan unit 20 are part of the functions prepared to solve the problems described above. The functions of these units will be described below.

[0057] The driver operation determination unit 25 checks the operation command values ​​sent from each of the control units 32, 33, and 34 to each of the actuators 7, 8, and 9, as well as the signals emitted when the driver operates the driving operation tool 50, to determine whether the driver has performed an intervention to change the braking force acting on the vehicle. More specifically, the driver operation determination unit 25 determines whether a braking operation has been performed. If it is determined that the driver has performed a braking operation, a driver operation signal is sent from the driver operation determination unit 25 to the target trajectory information update request unit 24.

[0058] The safety system operation determination unit 26 checks the operation command values ​​sent from each of the control units 32, 33, and 34 to each of the actuators 7, 8, and 9, as well as the signals emitted when the safety system 40 outputs an operation quantity, to determine whether the safety system 40 has performed an operation intervention that changes the braking force acting on the vehicle. If it is determined that an operation intervention has been performed through the operation of the safety system 40, the safety system operation determination unit 26 sends a safety system operation signal to the target trajectory information update request unit 24.

[0059] If a driver operation signal or a safety system operation signal is received, the target trajectory information update request unit 24 requests the target trajectory information generation unit 23 to update the target trajectory information. Updating the target trajectory information involves reconstructing the driving plan, and more specifically, reconstructing the speed plan included in the driving plan.

[0060] When a request is made to update the target trajectory information, the target trajectory information generation unit 23 reconstructs the speed plan based on the actual speed of the vehicle measured by the vehicle sensor 6. More specifically, the target trajectory information generation unit 23 corrects the planned speed to the current actual speed and resets the relationship between the position of each control point on the target path and the transit time at that position. When setting the planned acceleration for each control position in a time-related manner, the target trajectory information generation unit 23 also corrects the planned acceleration at each control point to the actual acceleration. (See reference...) Figure 4A and Figure 4B This describes an example of a method for reconstructing a driving plan from the target trajectory information generation unit 23.

[0061] Figure 4A This displays the target path, the vehicle's current position, and the vehicle's predicted position under normal driving conditions where the driver's braking operation or the operation of safety system 40 has not activated the braking. On the other hand, Figure 4B This diagram illustrates the target path, the vehicle's current position, and the predicted position of the vehicle during deceleration when braking is activated by the driver's braking operation or the operation of safety system 40. When the vehicle decelerates through braking, the predicted position deviates due to the change in speed. For example, suppose the driver performs a braking operation to brake the vehicle between time T0 and time T1. In this case, Figure 4A and Figure 4B The comparison indicates a deviation between the predicted position of the vehicle during deceleration at time T1 immediately following braking and the predicted position during normal operation.

[0062] In this embodiment, when an intervention to change the braking force acting on the vehicle is performed during autonomous driving, the planned speed determined by the speed plan is matched with the actual speed, and the relationship between control points and transit times on the target path is reset based on the actual speed. Therefore, the predicted position of the vehicle after a deceleration time T1 is correctly calculated. Then, control point P3, corresponding to the predicted position of the vehicle at the correctly calculated time T3, is determined as the reference point for time T1. Similarly, control point P4, corresponding to the predicted position of the vehicle at the correctly calculated time T4, is determined as the reference point for time T2. Reconstructing the driving plan in this way prevents a decrease in the vehicle's target path following ability and prevents unstable vehicle behavior, even when there is a difference between the planned and actual speeds. Interventions to change the braking force acting on the vehicle include, for example, braking operations performed by the driver and operations performed by safety systems. Safety systems include, for example, PCS, ABS, VSC, and TRC.

[0063] 3. Example of a driving plan reconstruction method

[0064] An example of the driving plan reconstruction method performed by the autonomous driving system 10 in this embodiment will be described by comparing the vehicle behavior between the autonomous driving system 10 in this embodiment and the autonomous driving system in the comparative example. First, reference will be made to... Figure 5 Examples to describe the behavior of vehicles in an autonomous driving system in the comparative examples. Figure 5 This is a time-series diagram showing the time variations of parameters related to the behavior of the vehicle in the autonomous driving system in the comparative example. From top to bottom, the following parameters are shown: braking force applied by the driver's braking action or the operation of the safety system; planned speed vs. actual speed; driving force applied by the autonomous driving system's speed control (ADS speed control); longitudinal acceleration of the vehicle; and reference curvature vs. the actual reference curvature value.

[0065] In the autonomous driving system of the comparative example, for instance, when the driver performs a braking operation and the braking force is applied to the vehicle, a difference arises between the planned speed and the actual speed. Since the speed control of the autonomous driving system includes feedback control, the driving force is increased by operating the drive actuators to reduce the difference between the planned and actual speeds. As a result, as indicated by problem 1 in the figure, braking force disturbances, for example, randomly generated by the driver's braking operation, are amplified by the driving force applied by the speed control of the autonomous driving system.

[0066] Subsequently, when the braking force suddenly decreases while the driving force and braking force are interfering with each other, the vehicle's longitudinal acceleration changes abruptly, as indicated by Problem 2 in the figure. This sudden change in longitudinal acceleration causes vehicle instability and discomfort to the occupants.

[0067] Furthermore, in the autonomous driving system of the comparative example, the steering control reference is based on the curvature determined by the assumption that the vehicle is traveling at a planned speed. Therefore, when the planned speed differs from the actual speed, an error occurs between the reference curvature of the steering control reference and the true value of the reference curvature, which is used as a reference, as indicated by question 3 in the figure. Similarly, errors occur between the true values ​​and the values ​​of the steering control reference in lateral deviation and yaw angle deviation. These errors lead to a reduction in the vehicle's ability to follow the target path and sudden steering maneuvers that cause unstable vehicle behavior.

[0068] Next, we will refer to Figure 6 Here is an example to describe the behavior of a vehicle controlled by the autonomous driving system 10 in this embodiment. Figure 6 This is a timing diagram showing the time variations of parameters related to the behavior of the vehicle in the autonomous driving system 10 in this embodiment. The following parameters are shown from top to bottom: braking force applied by the driver's braking operation or the operation of the safety system 40, planned speed and actual speed, driving force applied by the speed control of the autonomous driving system (ADS speed control), longitudinal acceleration of the vehicle, and reference curvature and the actual value of reference curvature.

[0069] When an operational intervention occurs, such as a driver's braking operation and braking force is applied to the vehicle, the autonomous driving system 10 in this embodiment reconstructs the speed plan. During speed plan reconstruction, the planned speed is matched to the actual speed, which, as indicated by Countermeasure 1, decreases due to the braking force when applied. Matching the planned speed to the actual speed in this way eliminates the difference between the planned and actual speeds and disables feedback control of the speed control. As a result, there is no interference between the braking force and the driving force applied by the speed control of the autonomous driving system, as indicated by Effect 1 in the figure. Note that when the planned speed is matched to the actual speed by reconstructing the driving plan, the integral or learned values ​​of the feedback control of steering control and speed control are reset or maintained.

[0070] Furthermore, since there is no difference between the planned speed and the actual speed, there is no difference between the reference curvature used for steering control and the true value of the reference curvature used as a reference, as indicated by effect 2 in the figure. The same applies to lateral deviation and yaw angle deviation. Referencing appropriate reference values ​​in steering control allows the vehicle's ability to follow the target path and the stability of its behavior to be maintained.

[0071] Methods for adjusting the feedback gain of steering control are known as approaches to improve a vehicle's target path following ability and behavioral stability. However, while increasing the feedback gain enhances the vehicle's target path following ability, it reduces behavioral stability. Conversely, decreasing the feedback gain stabilizes the vehicle's behavior but reduces its target path following ability. In contrast, the method described above for reconstructing the driving plan ensures both the vehicle's target path following ability and its behavioral stability.

[0072] Subsequently, when operational interventions, such as driver braking, cease and braking force is no longer applied to the vehicle, a speed plan is established so that the planned speed gradually increases from the actual speed at that time to the speed before the operational intervention, as indicated by countermeasure 2 in the figure. That is, a speed plan is established to gradually accelerate the vehicle. The gradual increase in planned speed correspondingly leads to a gradual increase in actual speed. This speed plan reduces the sudden change in longitudinal acceleration of the vehicle after a sudden decrease in braking force, as indicated by effect 3 in the figure, thereby stabilizing the vehicle's behavior.

[0073] One known method for reducing sudden changes in longitudinal acceleration is to gradually change the driving force. Changing the driving force gradually in this way prevents a sudden change in longitudinal acceleration after a sudden reduction in braking force; however, gradually changing the driving force makes it more difficult for the actual speed to quickly reach the planned speed. As a result, the problem caused by the difference between planned and actual speed remains unresolved. In contrast, the method described above for reconstructing the driving plan enables the reduction of the sudden change in longitudinal acceleration after a sudden reduction in braking force while simultaneously reducing the difference between planned and actual speed.

[0074] 4. Rebuild driving plan during ABS operation

[0075] Finally, an example of a method for reconstructing the driving plan when ABS is operated will be described below. There is a difference between the method for reconstructing the driving plan when ABS is operated and the method for reconstructing the driving plan when another safety system 40 is operated. Using TRC as an example of another safety system 40, the method for reconstructing the driving plan when ABS is operated and the method for reconstructing the driving plan when TRC is operated will be compared below.

[0076] Figure 7This diagram illustrates the relationship between the target acceleration determined by the autonomous driving control and the actual acceleration achieved through TRC operation. During normal operation when TRC is not running, the planned acceleration determined by the speed plan is set equal to the target acceleration. When TRC is activated and the vehicle is braked, the actual acceleration is maintained below the target acceleration, resulting in a change in the vehicle's speed. This causes the autonomous driving system 10 to begin reconstructing the driving plan. During the reconstruction of the driving plan during TRC operation, the speed plan is reconstructed such that the planned speed matches the actual speed, and the planned acceleration matches the actual acceleration. Reconstructing the driving plan in this way prevents a decrease in the vehicle's target path following ability and prevents unstable vehicle behavior.

[0077] on the other hand, Figure 8 This is a diagram illustrating the relationship between the target acceleration determined by autonomous driving control and the actual acceleration achieved through ABS operation. During normal operating time when ABS is not in operation, the planned acceleration determined by the speed plan is set to be equal to the target acceleration. Since ABS is operated when braking the vehicle via autonomous driving control, the target acceleration at this time is negative. When ABS is operated and the braking force acting on the vehicle decreases, the actual acceleration becomes less than the target acceleration, resulting in a change in the vehicle's speed. This causes the autonomous driving system 10 to begin reconstructing the driving plan.

[0078] During ABS operation, when reconstructing the driving plan, the speed plan is reconstructed such that the planned speed matches the actual speed, and the planned acceleration matches the target acceleration determined by the autonomous driving control. ABS is activated when the required deceleration cannot be achieved due to road surface limitations. If the planned acceleration matches the actual acceleration in such a situation, the required deceleration cannot be achieved when the road surface conditions are restored. Therefore, during ABS operation, the planned acceleration is kept matched with the target acceleration, not the actual acceleration. This prevents a decrease in the vehicle's target path following ability and prevents unstable vehicle behavior, while ensuring the maximum possible deceleration achievable under road surface limitations.

Claims

1. An autonomous driving system, comprising: Electronic control unit, the electronic control unit being configured to: Generate a target path represented by a set of control points in a predetermined coordinate system, and a speed plan specifying the transit time or transit speed at the control points on the target path. Autonomous driving control is used to enable the vehicle to travel along a target path. When the driver intervenes during autonomous driving by changing the braking force applied to the vehicle and altering the vehicle's actual speed, the speed plan is reconstructed based on the vehicle's actual speed by resetting the relationship between i) the position of each control point on the target path and ii) the transit time or transit speed at each control point. Specifically, the operational intervention alters the braking force acting on the vehicle, and the speed plan is rebuilt such that the planned speed of the speed plan established before the operational intervention is changed to match the actual speed occurring when the operational intervention is executed. The planned speed then gradually increases from the time the operational intervention ends until the vehicle is predicted to be on the target path at a predetermined passage time. The autonomous driving control is performed based on the reconstructed speed plan using actuators installed on the vehicle. The electronic control unit is configured to, when the operational intervention is an ABS operation, reconstruct the speed plan by matching the planned acceleration determined by the speed plan with the target acceleration determined by the autonomous driving control and by matching the planned speed determined by the speed plan with the actual speed.

2. The autonomous driving system according to claim 1, wherein, The electronic control unit is configured to reconstruct the speed plan based on the actual speed and actual acceleration of the vehicle when the operational intervention is performed.

3. The autonomous driving system according to claim 2, wherein, The electronic control unit is configured to, when the operational intervention is performed, reconstruct the speed plan by matching the planned acceleration determined by the speed plan with the actual acceleration.

4. The autonomous driving system according to claim 1, wherein, The relationship between the position of each control point on the target path (i) and the transit time or transit speed at the position of each control point on the target path (ii) is reset to maintain the target path generated before the intervention.

5. An autonomous driving method, comprising: The electronic control unit (ECU) generates a target path represented by a set of control points in a predetermined coordinate system and a speed plan specifying the transit time or transit speed at the control points on the target path. Autonomous driving control enables the vehicle to travel along a target path. During autonomous driving, when the driver intervenes to change the braking force applied to the vehicle and alter its actual speed, the speed plan is reconstructed using the ECU based on the vehicle's actual speed by resetting the relationship between i) the position of each control point on the target path and ii) the transit time or transit speed at each control point's position. Specifically, the operational intervention alters the braking force acting on the vehicle, and the speed plan is rebuilt such that the planned speed of the speed plan established before the operational intervention is changed to match the actual speed occurring when the operational intervention is executed. The planned speed then gradually increases from the time the operational intervention ends until the vehicle is predicted to be on the target path at a predetermined passage time. The autonomous driving control is performed based on the reconstructed speed plan using actuators installed on the vehicle. The electronic control unit is configured to, when the operational intervention is an ABS operation, reconstruct the speed plan by matching the planned acceleration determined by the speed plan with the target acceleration determined by the autonomous driving control and by matching the planned speed determined by the speed plan with the actual speed.

6. The autonomous driving method according to claim 5 further includes: When the operational intervention is performed, the speed plan is reconstructed based on the vehicle's actual speed and actual acceleration.

7. The autonomous driving method according to claim 6 further includes: When the operational intervention is performed, the velocity plan is reconstructed by matching the planned acceleration determined by the velocity plan with the actual acceleration.

8. The autonomous driving method according to claim 5, wherein, The relationship between the position of each control point on the target path (i) and the transit time or transit speed at the position of each control point on the target path (ii) is reset to maintain the target path generated before the intervention.

Images