Vehicle control method, vehicle-mounted control equipment, vehicle and storage medium

Abstract

The embodiment of the invention provides a vehicle control method, vehicle-mounted control equipment, a vehicle and a storage medium. The method comprises the steps that current state data, target state data and current scene data corresponding to the vehicle are acquired; determining a feed-forward compensation amount based on the target state data; processing the current state data and the target state data by adopting a state space model constructed based on a vehicle dynamics model, and determining a feedback control quantity; determining an integral compensation amount based on the current state data, the target state data and the current scene data; and determining a comprehensive control instruction according to the feedforward compensation amount, the feedback control amount and the integral compensation amount, and controlling the target component to work according to the comprehensive control instruction. The method is used for rapidly and accurately controlling the vehicle so as to improve the response performance in the automatic driving process of the vehicle and the tracking accuracy of the preset trajectory.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of autonomous driving, and in particular to a vehicle control method, a vehicle-mounted control device, a vehicle and a storage medium. BACKGROUND

[0002] In the technical field of autonomous driving, a preset trajectory is usually planned in advance, and a vehicle is controlled to track the preset trajectory for driving. In the prior art, the vehicle is usually controlled in real time according to a tracking error of the vehicle (for example, an error between an actual speed of a current position and a preset speed, or an error between an actual position and a preset position, etc.) to control the vehicle to automatically follow the preset trajectory. However, since different working conditions exist in an actual scene, the tracking error of the vehicle is usually unable to timely capture real-time working conditions of the actual scene, and the control of the vehicle is relatively lagging, which leads to a decrease in the control accuracy of the vehicle, and further leads to a decrease in the response performance of the vehicle in the process of autonomous driving and a decrease in the accuracy of the vehicle in tracking the preset trajectory. Therefore, how to quickly and accurately control the vehicle to improve the response performance of the vehicle in the process of autonomous driving and the accuracy of the vehicle in tracking the preset trajectory is a technical problem to be solved at present. SUMMARY

[0003] Embodiments of the present application provide a vehicle control method, a vehicle-mounted control device, a vehicle and a storage medium, and aim to solve the technical problem of how to quickly and accurately control the vehicle to improve the response performance of the vehicle in the process of autonomous driving and the accuracy of the vehicle in tracking the preset trajectory.

[0004] A vehicle control method comprises: acquiring current state data, target state data and current scene data corresponding to a vehicle, the current state data being state data collected at a current position, the target state data being preset state data corresponding to a target tracking point, and the current scene data being scene data in which the vehicle travels at the current position; determining a feedforward compensation amount based on the target state data; processing the current state data and the target state data by using a state space model constructed based on a vehicle dynamics model to determine a feedback control amount; determining an integral compensation amount based on the current state data, the target state data and the current scene data; determining a comprehensive control instruction according to the feedforward compensation amount, the feedback control amount and the integral compensation amount, and controlling a target component to work according to the comprehensive control instruction.

[0005] In the embodiment, the feedback control amount is determined according to the target state data of the target tracking point, the target component can be driven according to the feedback control amount before the error at the current position is generated, the time delay of the vehicle control is shortened, the vehicle is quickly controlled, and the response speed performance in the automatic driving process of the vehicle is improved. The state space model constructed based on the vehicle dynamics model is used to process the current state data and the target state data, and the feedback control amount is determined, so as to ensure the continuity and stability of the vehicle control according to the feedback control amount. Based on the current state data, the target state data and the current scene data, the integral compensation amount is determined, the current scene data reflecting the actual emergency in the current scene is considered, the error corresponding to the feedback control amount is compensated, the error is reduced, the vehicle is quickly and accurately controlled, and the speed of controlling the vehicle and the accuracy of the vehicle tracking the preset trajectory are improved. The method can quickly and accurately control the vehicle according to the comprehensive control instruction determined by the feedforward compensation amount, the feedback control amount and the integral compensation amount under the condition of ensuring the continuity and stability of the vehicle control, improve the response speed of controlling the vehicle, and improve the accuracy of the vehicle tracking the preset trajectory.

[0006] Preferably, the target state data comprises a target acceleration corresponding to the target tracking point. The feedforward compensation amount is determined based on the target acceleration and a pre-labeled feedforward compensation parameter.

[0007] In the embodiment, the product of the feedforward compensation parameter and the target acceleration of the target tracking point is determined as the feedforward compensation amount, a "open loop" driving instruction is generated in advance according to the target acceleration, which is used to effectively shorten the delay of the initial response of the vehicle control system, and the target component is driven to execute the control instruction corresponding to the feedforward compensation amount before the error is actually generated, thereby providing a "first driving force" for the fast dynamic response.

[0008] Preferably, the current state data comprises a current longitudinal speed and a current longitudinal position corresponding to the current position; and the target state data comprises a target longitudinal speed and a target longitudinal position corresponding to the target tracking point. The current state data and the target state data are processed by using the state space model constructed based on the vehicle dynamics model, and the feedback control amount is determined, including: A target function is determined based on the current longitudinal speed and the current longitudinal position corresponding to the current position, and the target longitudinal speed and the target longitudinal position corresponding to the target tracking point. A constraint condition is determined based on a control input amplitude limit and a control input rate limit in the state space model. roll out the state space model based on the constraint condition, and determine a target control sequence when the target function is minimized, the target control sequence being a sequence of control inputs of a plurality of preset tracking points in a preset trajectory corresponding to a future time period; the target tracking point being a first preset tracking point in the preset trajectory; determine a feedback control amount corresponding to the target tracking point according to the target control sequence.

[0009] In this embodiment, the state space model is rolled out based on the target function and the constraint condition to obtain a feedback control amount that can accurately control the vehicle, and the vehicle can accurately and stably track the target tracking point through the feedback control amount.

[0010] Preferably, the integral compensation amount is determined based on the current state data, the target state data and the current scene data, comprising: when at least one of the target state data and the current scene data satisfies a dynamic scene condition, the integral compensation amount is determined based on the current state data and the target state data; when neither the target state data nor the current scene data satisfies the dynamic scene condition, the integral compensation amount is determined as 0.

[0011] In this embodiment, the integral compensation amount is determined according to whether the target state data and the current scene data satisfy the dynamic scene condition, so as to compensate the feedback control amount according to the integral compensation amount, reduce the error, and accurately control the vehicle to ensure the accuracy of the vehicle tracking the preset trajectory.

[0012] Preferably, the current scene data comprises a planned behavior signal; and the target state data comprises a target acceleration and a target jerk corresponding to the target tracking point. The dynamic scene condition comprises a planned behavior enabling condition and a target state enabling condition. The planned behavior enabling condition comprises that the planned behavior signal is an emergency signal, and the emergency signal comprises an emergency acceleration signal and an emergency deceleration signal. The target state enabling condition comprises that an absolute value of the target acceleration is greater than a preset acceleration threshold, and an absolute value of the target jerk is greater than a preset jerk threshold.

[0013] In the embodiment, whether the dynamic scene planning condition is met is determined according to the planning behavior signal, whether the planning behavior enabling condition is met is determined according to the target acceleration and the target jerk, and whether the dynamic scene condition is met is determined according to the dynamic scene planning condition and the planning behavior enabling condition, so as to determine the integral compensation amount according to whether the dynamic scene condition is met, compensate the feedback control amount, reduce the error, improve the accuracy of vehicle control, and further improve the accuracy of the vehicle in tracking the preset trajectory.

[0014] Preferably, the current state data includes a current speed corresponding to the current position, and the target state data includes a target speed corresponding to the target tracking point. The determining of the integral compensation amount according to the current state data and the target state data comprises: When at least one of the target state data and the current scene data meets the dynamic scene condition, an enabling signal value is determined. A target integral value is determined by integrating a speed difference value corresponding to a target time period, wherein the speed difference value is a difference between the target speed and the current speed, and the target time period is a time interval corresponding to the current position and the target tracking point on the preset trajectory. The initial compensation amount is determined by correcting the target integral value using the enabling signal and an integral gain coefficient. The integral compensation amount is determined by processing the initial compensation amount according to an integral compensation upper limit and an integral compensation lower limit.

[0015] In the embodiment, the target integral value is determined by integrating the difference between the target speed and the current speed, and the integral compensation amount is determined according to the target integral value, which can effectively determine the difference between the current speed and the target speed, determine a more accurate integral compensation amount, and process the initial compensation amount according to the integral compensation upper limit and the integral compensation lower limit, thereby avoiding insufficient compensation, excessive error, and further insufficient control accuracy, and avoiding excessive compensation, overshoot and oscillation, and affecting the smoothness of vehicle control.

[0016] Preferably, the comprehensive control instruction is a sum of the feedforward compensation amount, the feedback control amount and the integral compensation amount.

[0017] In the embodiment, the feedforward compensation amount, the feedback control amount and the integral compensation amount are comprehensively considered to obtain a comprehensive control instruction for controlling the vehicle to follow the preset trajectory. In the case of ensuring the continuity and stability of vehicle control according to the feedback control amount, the vehicle can be quickly and accurately controlled according to the feedforward compensation amount to improve the response speed of the vehicle, and the error can be compensated according to the integral compensation amount to improve the accuracy of vehicle control, and further improve the accuracy of the vehicle in tracking the preset trajectory.

[0018] A vehicle-mounted control device, comprising a processor and a memory, wherein, The memory is configured to store a computer program. The processor is configured to execute the program stored in the memory to implement the vehicle control method.

[0019] A vehicle comprising the vehicle-mounted control device.

[0020] A computer-readable storage medium, the computer-readable storage medium stores a computer program, the computer program is executed by a processor to implement the vehicle control method. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 is a flowchart of a vehicle control method provided by an embodiment of the present application; Figure 2 is another flowchart of a vehicle control method provided by an embodiment of the present application; Figure 3 is another flowchart of a vehicle control method provided by an embodiment of the present application; Figure 4 is another flowchart of a vehicle control method provided by an embodiment of the present application; Figure 5 is a structural diagram of a vehicle-mounted control device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the technical problems, technical solutions and beneficial effects solved by the present application clearer, the present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0023] A vehicle control method provided by an embodiment of the present application comprises: acquiring current state data, target state data and current scene data corresponding to a vehicle, the current state data being state data collected at a current position, the target state data being state data of a target tracking point in a preset trajectory, and the current scene data being scene data in which the vehicle travels at the current position; determining a feedforward compensation amount based on the target state data; processing the current state data and the target state data by using a state space model constructed based on a vehicle dynamics model to determine a feedback control amount; determining an integral compensation amount based on the current state data, the target state data and the current scene data; determining a comprehensive control instruction according to the feedforward compensation amount, the feedback control amount and the integral compensation amount, and controlling a target component to work according to the comprehensive control instruction. The method is used to quickly and accurately control the vehicle, so as to improve the response performance in the automatic driving process of the vehicle and the accuracy of tracking the preset trajectory.

[0024] like Figure 1 As shown, this application provides a vehicle control method, which is applied to... Figure 5 Taking the vehicle-mounted control equipment as an example, the explanation includes the following steps: S101: Obtain the current status data, target status data, and current scene data corresponding to the vehicle. The current status data is the status data collected at the current location, the target status data is the status data of the target tracking point in the preset trajectory, and the current scene data refers to the scene data of the vehicle driving at the current location. S102: Determine the feedforward compensation amount based on the target state data; S103: Using a state-space model built on a vehicle dynamics model, the current state data and target state data are processed to determine the feedback control quantity; S104: Determine the integral compensation amount based on the current state data, target state data, and current scene data; S105: Determine the integrated control command based on the feedforward compensation, feedback control, and integral compensation, and control the target component to work according to the integrated control command.

[0025] The current state data refers to the state data collected at the current location. The target state data refers to the state data of the target tracking point in the preset trajectory. The target tracking point refers to the tracking point set in the preset trajectory in the vehicle's autonomous driving scenario. The preset trajectory refers to the preset trajectory that the vehicle needs to follow. The target tracking point refers to the preset tracking point in the preset trajectory that has the closest longitudinal distance to the current location. The preset tracking point refers to the tracking point set in the preset trajectory at a time interval corresponding to a certain preset duration. In this embodiment, in a preset trajectory, at each... The time interval is set by establishing a preset tracking point, and multiple preset tracking points are determined to facilitate vehicle control. The time interval reaches a preset tracking point to achieve the purpose of tracking the preset trajectory. This is the preset duration. The longitudinal direction is the direction of vehicle travel. The longitudinal distance is the distance along the longitudinal direction; for example, the current location's longitudinal coordinates are... The preset trajectory contains M preset tracking points, where the coordinates of the i-th preset tracking point in the vertical direction are... ,Will The coordinates of each preset tracking point in the longitudinal direction among the M target tracking points. Distance calculations are performed to determine the preset tracking point j with the smallest distance, and this preset tracking point j is then designated as the target tracking point corresponding to the current position. Current scene data refers to the scene data of the vehicle traveling at its current position.

[0026] As an example, in step S101, the vehicle-mounted control device acquires the current state data and the current scene data corresponding to the current position collected by the data collection device in real time, determines the target tracking point closest to the current position in the longitudinal direction on the preset trajectory according to the current state data, and queries the target state data corresponding to the target tracking point in the preset trajectory stored in the system database. Understandably, during the automatic driving of the vehicle according to the preset trajectory, a plurality of preset tracking points are usually set on the preset trajectory, so as to control the vehicle to move following the preset tracking points and realize the following of the preset trajectory.

[0027] Among them, the feedforward compensation amount refers to the data capable of controlling the vehicle determined according to the target state data.

[0028] As an example, in step S102, after the vehicle-mounted control device determines the target state data corresponding to the target tracking point, the target state data corresponding to the target tracking point is directly processed to determine the feedforward compensation amount for controlling the vehicle at the target tracking point, without the need to process the data of the current position. The feedforward compensation amount determined by the method can provide a driving control amount for vehicle control, which is used to drive the target component before the error at the current position occurs, shortens the time delay of vehicle control, so as to quickly control the vehicle and improve the response speed performance in the automatic driving process of the vehicle.

[0029] Among them, the feedback control amount refers to the control amount determined by using the state space model.

[0030] As an example, in step S103, the vehicle-mounted control device establishes a state space model based on the vehicle dynamics model, for example, a model predictive control model (MPC), and performs rolling solution on the state space model within a future time period according to the current state data corresponding to the current position and the target state data corresponding to the target tracking point, to predict the control input corresponding to N preset tracking points within the future time period. The first control input in the N control inputs is determined as the feedback control amount, so as to subsequently determine the comprehensive control instruction according to the feedback control amount, and guarantee the stability and safety in the automatic driving process of the vehicle.

[0031] Among them, the integral compensation amount is used to compensate the error of the feedback control amount according to the current scene data.

[0032] As an example, in step S104, the vehicle-mounted control device determines the integral compensation amount corresponding to the current scene data after determining the current scene data corresponding to the current position, and realizes error compensation of the feedback control amount based on the integral compensation amount by performing integral control on the current state data and the target state data according to the current scene data. Understandably, when the state space model is used to predict the control input corresponding to the target tracking point, the error between the current state data and the target state data is usually used for prediction to ensure the continuity and smoothness of the vehicle control in the time domain. However, when an emergency situation exists in the current actual scene, the continuity of the vehicle control in the time domain is usually interrupted, and the state space model does not consider the situation in the current actual scene, which usually causes a lag in the vehicle control and increases the error, resulting in a decrease in the tracking accuracy. In the example, the current scene data is obtained in real time at the current position to reflect various emergency situations existing in the current actual scene through the current scene data. The integral control is performed on the current state data and the target state data according to the current scene data to determine the integral compensation amount corresponding to the current scene data, compensate for the error of the feedback control amount caused by various emergency situations existing in the current actual scene, reduce the error of the feedback control amount, and improve the accuracy of the vehicle tracking the preset trajectory.

[0033] The comprehensive control instruction is a control input determined by comprehensively considering the speed and accuracy of the vehicle control. The target component is a component that executes the comprehensive control instruction.

[0034] As an example, in step S105, the vehicle-mounted control device processes the feedforward compensation amount, the feedback control amount, and the integral compensation amount according to a preset method to determine a comprehensive control instruction for controlling the vehicle, so that the target component of the vehicle controls the vehicle quickly and accurately according to the comprehensive control instruction, and the speed of controlling the vehicle and the accuracy of the vehicle tracking the preset trajectory are improved while ensuring the continuity and smoothness of the vehicle control.

[0035] In the embodiment, the feedback control quantity is determined according to the target state data of the target tracking point, the target component can be driven according to the feedback control quantity before the error at the current position is generated, the time delay of the vehicle control is shortened, the vehicle is quickly controlled, and the response speed performance in the automatic driving process of the vehicle is improved. The state space model constructed based on the vehicle dynamics model is used to process the current state data and the target state data, and the feedback control quantity is determined, so as to ensure the continuity and stability of the vehicle control according to the feedback control quantity. Based on the current state data, the target state data and the current scene data, the integral compensation quantity is determined, the current scene data reflecting the actual emergency in the current scene is considered, the error corresponding to the feedback control quantity is compensated, the error is reduced, the vehicle is quickly and accurately controlled, and the speed of controlling the vehicle and the accuracy of the vehicle tracking the preset trajectory are improved. The method can quickly and accurately control the vehicle according to the comprehensive control instruction determined by the feedforward compensation quantity, the feedback control quantity and the integral compensation quantity under the condition of ensuring the continuity and stability of the vehicle control, improve the response speed of controlling the vehicle, and improve the accuracy of the vehicle tracking the preset trajectory.

[0036] In an embodiment, the target state data includes a target acceleration corresponding to the target tracking point. The feedforward compensation quantity is determined based on the target acceleration and a pre-labeled feedforward compensation parameter.

[0037] The target acceleration refers to a preset acceleration corresponding to the target tracking point. The feedforward compensation parameter refers to a pre-labeled parameter.

[0038] As an example, the vehicle-mounted control device obtains a target acceleration corresponding to a target tracking point closest to the current position in the longitudinal direction , and determines the product of the pre-labeled feedforward compensation parameter and the target acceleration corresponding to the target tracking point as the feedforward compensation quantity . That is = .

[0039] In the embodiment, the product of the feedforward compensation parameter and the target acceleration of the target tracking point is determined as the feedforward compensation quantity, a "open loop" driving instruction is generated in advance according to the target acceleration, which is used to effectively shorten the delay of the initial response of the vehicle control system, and the target component is driven in advance to execute the control instruction corresponding to the feedforward compensation quantity before the error is actually generated, thereby providing "first driving force" for fast dynamic response.

[0040] In an embodiment, the current state data includes a current longitudinal speed and a current longitudinal position corresponding to the current position; and the target state data includes a target longitudinal speed and a target longitudinal position corresponding to the target tracking point. As shown in Figure 2 Step S103, i.e. using the state space model constructed based on the vehicle dynamics model, processes the current state data and the target state data to determine the feedback control amount, including: S201: determining a target function based on the current longitudinal speed and the current longitudinal position corresponding to the current position, and the target longitudinal speed and the target longitudinal position corresponding to the target tracking point; S202: determining a constraint condition based on the control input amplitude limit and the control input rate limit in the state space model; S203: performing a rolling solution on the state space model based on the constraint condition, and determining a target control sequence when the target function is minimized, the target control sequence being a sequence formed by the control inputs of the plurality of preset tracking points in the preset trajectory corresponding to the future time period; the target tracking point being the first preset tracking point in the preset trajectory; S204: determining the feedback control amount corresponding to the target tracking point according to the target control sequence.

[0041] Wherein, the current longitudinal speed refers to the speed of the vehicle in the longitudinal direction at the current position. The current longitudinal position refers to the position of the vehicle in the longitudinal direction at the current position. The target longitudinal speed refers to the preset speed of the vehicle in the longitudinal direction at the target tracking point. The target longitudinal position refers to the preset position of the vehicle in the longitudinal direction at the target tracking point. In this example, the longitudinal direction of the vehicle is the direction of travel along the vehicle body. For example, for the vehicle body, the coordinate of the longitudinal direction is the horizontal coordinate Wherein, i refers to the i-th preset tracking point in the M preset tracking points, and in particular, i=0 represents the current position.

[0042] As an example, in step S201, the vehicle-mounted control device processes the current longitudinal speed and the current longitudinal position corresponding to the current position, and the target longitudinal speed and the target longitudinal position corresponding to the target tracking point to determine a target function capable of predicting the control input corresponding to the plurality of preset tracking points in the future time period in the preset trajectory. Understandably, since the preset tracking point is a tracking point in the preset trajectory, and the target tracking point is the tracking point closest to the current position in the longitudinal direction, the target tracking point is the first preset tracking point in the future time period.

[0043] Wherein, the control input refers to the control amount of the vehicle controlled by the accelerator or the brake. The control input amplitude limit refers to the limit range of the size of the amplitude of the control input. The control input rate limit refers to the limit range of the size of the rate of change of the control input.

[0044] As an example, in step S202, the vehicle-mounted control device determines the control input amplitude limit and the control input rate limit in the state space model as the constraint condition. The control input amplitude limit is: The control input rate limit is: , is the control input of the i+1th preset tracking point, is the rate of change of the control input of the i+1th preset tracking point, is the minimum value of the control input, is the maximum value of the control input. is the minimum value of the rate of change of the control input, is the maximum value of the rate of change of the control input. Understandably, an input control input is input at the i-th preset tracking point, which is used to control the i+1th preset tracking point, so that is the control input of the i+1th preset tracking point, is the rate of change of the control input of the i+1th preset tracking point. In this example, the constraint condition of the state space model is set so that the state space model has feasibility in subsequent solving according to the constraint condition.

[0045] The target control sequence refers to the sequence of control inputs corresponding to each preset tracking point predicted in the solving process of the constraint condition and the target function.

[0046] As an example, in step S203, the vehicle-mounted control device solves the state space model by rolling solving in the future time period, and determines the target control sequence corresponding to the control input of each preset tracking point in the future time period when the target function is minimized and the control input meets the preset condition , completing the solving of the state space model.

[0047] As an example, in step S204, the vehicle-mounted control device determines the first control input in the target control sequence as the feedback control amount. Understandably, the first preset tracking point is the target tracking point, and the target tracking point is the preset tracking point closest to the current position in the longitudinal direction, represents the control input for controlling the target tracking point at the current position. The first control input in the target control sequence obtained by rolling solving the constraint condition and the target function in the future time period is close to the input of the target tracking point, so that the first control input in the target control sequence is determined as the feedback control amount corresponding to the target tracking point, which can accurately control the vehicle at the current position and make the vehicle accurately track the target tracking point.

[0048] In this embodiment, the state space model is solved according to the target function and the constraint condition to obtain a feedback control quantity capable of accurately controlling the vehicle, and the vehicle can accurately and stably track the target tracking point through the feedback control quantity.

[0049] In an embodiment, the target function is determined based on a current error term corresponding to the current position and an error function term corresponding to the target tracking point; The current error term is determined based on a longitudinal position error corresponding to the current position and a longitudinal velocity error corresponding to the current position; The longitudinal position error corresponding to the current position is determined based on a current longitudinal position corresponding to the current position and a reference longitudinal position corresponding to the current position, and the longitudinal velocity error corresponding to the current position is determined based on a current longitudinal velocity corresponding to the current position and a reference longitudinal velocity corresponding to the current position; The error function term is determined based on longitudinal position errors corresponding to a plurality of preset tracking points and longitudinal velocity errors corresponding to the plurality of preset tracking points in a future time period, wherein the first preset tracking point is the target tracking point; The longitudinal position error corresponding to the preset tracking point is determined based on a predicted longitudinal position corresponding to the preset tracking point and a reference longitudinal position, and the reference longitudinal position corresponding to the target tracking point is a target longitudinal position; The longitudinal velocity error corresponding to the preset tracking point is determined based on a predicted longitudinal velocity corresponding to the preset tracking point and a reference longitudinal velocity, and the reference longitudinal velocity corresponding to the target tracking point is a target longitudinal velocity.

[0050] As an example, in step S201, the vehicle-mounted control device determines the state space model as The longitudinal velocity error and the longitudinal position error corresponding to the current position, and the target longitudinal velocity and the target longitudinal position corresponding to the target tracking point are processed to determine the target function. Wherein, is the longitudinal position error corresponding to the i+1th preset tracking point, is the longitudinal velocity error corresponding to the i+1th preset tracking point, is the control input input at the i preset tracking point for controlling the i+1th preset tracking point. When i=0, is the longitudinal position error corresponding to the current position, is the longitudinal velocity error corresponding to the current position, is the control input input at the current position for controlling the target tracking point (i.e. the first preset tracking point in the preset trajectory). In this example, the target function is determined by the following two methods: Method one, the target function is wherein N is the number of preset tracking points in the future time period in the preset trajectory, and N is is the current error term, i≠0, is the error function term.

[0051] wherein = wherein is the longitudinal position error corresponding to the current position, is the longitudinal velocity error corresponding to the current position.

[0052] wherein is the current longitudinal position corresponding to the current position, is the reference longitudinal position corresponding to the current position, the current longitudinal position corresponding to the current position being the actual longitudinal position in the current position, and the reference longitudinal position corresponding to the current position being the preset longitudinal position of the vehicle at the current time. wherein is the current longitudinal velocity corresponding to the current position, is the reference longitudinal velocity corresponding to the current position, the current longitudinal velocity corresponding to the current position being the actual longitudinal velocity of the vehicle at the current position, and the reference longitudinal velocity corresponding to the current position being the preset longitudinal velocity of the vehicle at the current position.

[0053] wherein is the longitudinal position error corresponding to the i-th preset tracking point, is the longitudinal velocity error corresponding to the i-th preset tracking point, wherein is the current longitudinal position corresponding to the i-th preset tracking point, is the reference longitudinal position corresponding to the i-th preset tracking point, the current longitudinal position corresponding to the i-th preset tracking point being the longitudinal position of the i-th preset tracking point predicted by means of rolling solution in the future time period, and the reference longitudinal position corresponding to the i-th preset tracking point being the preset longitudinal position that the i-th preset tracking point should have. In particular, when i=1, the first preset tracking point is the target tracking point, and the reference longitudinal position corresponding to the first preset tracking point is the target longitudinal position corresponding to the target tracking point. wherein is the current longitudinal velocity corresponding to the i-th preset tracking point, ​​​is a reference longitudinal velocity corresponding to the ith preset tracking point. A current longitudinal velocity corresponding to the ith preset tracking point is a longitudinal velocity of the ith preset tracking point predicted by means of rolling solution in a future time period, and the current longitudinal velocity corresponding to the ith preset tracking point is a preset longitudinal velocity that should be possessed by the ith preset tracking point. In particular, when i = 1, the first preset tracking point is the target tracking point, and the reference longitudinal velocity corresponding to the first preset tracking point is the target longitudinal velocity corresponding to the target tracking point.

[0054] Method two, the objective function is wherein, is a control input corresponding to the ith preset tracking point, when i = 0, is a control input input at the current position for controlling the target tracking point. In this example, the control input is acceleration.

[0055] In this embodiment, the target function is determined according to the longitudinal velocity error and the longitudinal position error corresponding to the current position, and the longitudinal position error corresponding to the plurality of preset tracking points and the longitudinal velocity error corresponding to the plurality of preset tracking points in the future time period, so as to accurately solve the state space model according to the target function.

[0056] In an embodiment, as shown in Figure 3 Step S104, i.e., determining the integral compensation amount based on the current state data, the target state data and the current scene data, comprises: S301: When at least one of the target state data and the current scene data satisfies the dynamic scene condition, determining the integral compensation amount according to the current state data and the target state data; S302: When neither the target state data nor the current scene data satisfies the dynamic scene condition, determining the integral compensation amount as 0.

[0057] The dynamic scene condition is a condition for judging whether the integral compensation amount needs to be determined.

[0058] As an example, in step S301, the vehicle-mounted control device determines that there is an emergency in the current scene when it is determined that the target state data satisfies the dynamic scene condition, the current scene data satisfies the dynamic scene condition, or the target state data and the current scene data satisfy the dynamic scene condition, in order to ensure the accuracy of the control vehicle tracking the preset trajectory, the integral compensation amount is determined according to the current state data corresponding to the current position and the target state data corresponding to the target tracking point, for compensating the feedback control amount.

[0059] As an example, in step S302, the vehicle-mounted control device determines that there is no emergency in the current scene when it is determined that neither the target state data nor the current scene data satisfies the dynamic scene condition, and determines that the integral compensation amount is 0 without compensating the feedback control amount.

[0060] In this embodiment, the integral compensation amount is determined according to whether the target state data and the current scene data satisfy the dynamic scene condition, so as to compensate the feedback control amount according to the integral compensation amount, reduce the error, and accurately control the vehicle to ensure the accuracy of the vehicle tracking the preset trajectory.

[0061] In an embodiment, the current scene data includes a planning behavior signal; and the target state data includes a target acceleration and a target jerk corresponding to a target tracking point. The dynamic scene condition includes a planning behavior enabling condition and a target state enabling condition. The planning behavior enabling condition includes that the planning behavior signal is an emergency signal, and the emergency signal includes an emergency acceleration signal and an emergency deceleration signal. The target state enabling condition includes that an absolute value of the target acceleration is greater than a preset acceleration threshold, and an absolute value of the target jerk is greater than a preset jerk threshold.

[0062] The planning behavior signal is a signal for reflecting whether there is an emergency in the current scene. The emergency includes but is not limited to an emergency braking situation and an emergency acceleration situation. The target acceleration is an acceleration that the vehicle needs to reach at a target tracking point corresponding to a current position. The target jerk is a change amount of the target acceleration of the vehicle at the target tracking point. The planning behavior enabling condition is a condition for determining whether the integral compensation amount needs to be determined according to the planning behavior signal. The target state enabling condition is a condition for determining whether the integral compensation amount needs to be determined according to the target state data. The emergency signal is a signal for representing an emergency, such as an emergency acceleration signal and an emergency deceleration signal. The emergency acceleration signal is a signal required for an emergency acceleration situation of the vehicle. The emergency deceleration signal is a signal required for an emergency deceleration situation of the vehicle. The preset acceleration threshold is a preset threshold for judging the size of the target acceleration. The preset jerk threshold is a preset threshold for judging the size of the target jerk.

[0063] As an example, the vehicle-mounted control device obtains the planning behavior signal in the current scene data, determines that the planning behavior signal is an emergency signal when it is determined that the planning behavior signal is an emergency acceleration signal or an emergency deceleration signal, and determines that the planning behavior signal satisfies the planning behavior enabling condition.

[0064] The vehicle-mounted control device acquires a target acceleration and a target jerk contained in target state data, and determines that the target state data satisfies a planning behavior enabling condition when it is determined that an absolute value of the target acceleration is greater than a preset acceleration threshold value and an absolute value of the target jerk is greater than a preset jerk threshold value.

[0065] The vehicle-mounted control device determines that at least one of the target state data and the current scene data satisfies a dynamic scene condition when it is determined that one of the following three conditions is met: the planning behavior signal satisfies the planning behavior enabling condition and the target state data does not satisfy the planning behavior enabling condition; the planning behavior signal does not satisfy the planning behavior enabling condition and the target state data satisfies the planning behavior enabling condition; or the planning behavior signal satisfies the planning behavior enabling condition and the target state data satisfies the planning behavior enabling condition, and further performs step S301 of determining an integral compensation amount according to the current state data of the current position and the target state data of the target tracking point. Understandably, when the planning behavior signal is an emergency signal and / or the absolute values of the target acceleration and the target jerk are large, it indicates that there is a situation that requires an emergency change in the motion state in the current scene. Due to dynamic lag and model mismatch of control, etc., the error cannot be eliminated in time and will continue to exist, resulting in a decrease in the accuracy of vehicle control, affecting the smoothness of control, and reducing the accuracy of the vehicle tracking the preset trajectory. Therefore, when it is determined that there is a situation that requires an emergency change in the motion state in the scene, the integral compensation amount needs to be determined to compensate for the feedback control amount, reduce the error, and improve the accuracy of vehicle control.

[0066] In this embodiment, whether the dynamic scene planning condition is satisfied is determined according to the planning behavior signal, whether the planning behavior enabling condition is satisfied is determined according to the target acceleration and the target jerk, and whether the dynamic scene condition is satisfied is determined according to the dynamic scene planning condition and the planning behavior enabling condition, so as to determine the integral compensation amount according to whether the dynamic scene condition is satisfied, compensate for the feedback control amount, reduce the error, improve the accuracy of vehicle control, and further improve the accuracy of the vehicle tracking the preset trajectory.

[0067] In an embodiment, the current state data includes a current speed corresponding to the current position, and the target state data includes a target speed corresponding to the target tracking point.

[0068] The current speed refers to the actual speed of the vehicle at the current position. The target speed refers to the preset speed of the vehicle corresponding to the target tracking point. Understandably, to control the vehicle to accurately track the preset trajectory, a plurality of preset tracking points are usually set on the preset trajectory, and each preset tracking point corresponds to a preset speed. The target tracking point is the preset tracking point closest to the current position in the longitudinal direction, and the target tracking point also corresponds to a preset speed, i.e., the target speed.

[0069] In an embodiment, as shown in Figure 4 Step S301, i.e., determining the integral compensation amount according to the current state data and the target state data, comprises: S401: determining an enable signal value when at least one of the target state data and the current scene data satisfies the dynamic scene condition; S402: integrating the speed difference value corresponding to the target time period to determine a target integral value; the speed difference value is the difference between the target speed and the current speed; the target time period is the time interval corresponding to the current position and the target tracking point on the preset trajectory; S403: correcting the target integral value using the enable signal and the integral gain coefficient to determine an initial compensation amount; S404: processing the initial compensation amount according to the upper limit of the integral compensation and the lower limit of the integral compensation to determine the integral compensation amount.

[0070] The enable signal value is a signal value used to represent the need to determine the integral compensation amount.

[0071] As an example, in step S401, the vehicle-mounted control device determines the enable signal value to be 1 when at least one of the target state data and the current scene data satisfies the dynamic scene condition, which is used to start the integrator to determine the integral compensation amount. The vehicle-mounted control device determines the enable signal value to be 0 when neither the target state data nor the current scene data satisfies the dynamic scene condition, which is used to freeze the integrator to determine the integral compensation amount to be 0.

[0072] The target integral value refers to the integral value corresponding to the difference between the target speed and the current speed.

[0073] As an example, in step S402, the vehicle-mounted control device integrates the difference between the target speed corresponding to the target tracking point and the current speed corresponding to the current position over the target time period corresponding to the current position and the target tracking point to determine the target integral value when the enable signal value is 1, i.e., the target integral value is: wherein, is the target speed, is the current speed.

[0074] The integral gain signal is a signal used to correct the target integral value, which is pre-labeled. The initial compensation amount is the initially determined integral compensation amount.

[0075] As an example, in step S403, the vehicle-mounted control device corrects the target integral value using the enable signal and the integral gain coefficient to determine the initial compensation amount . Wherein, = SwichValue K , SwichValue is an enabling signal, and K is an integral gain coefficient.

[0076] wherein the integral compensation upper limit refers to the maximum value of the integral compensation amount, and the integral compensation lower limit refers to the minimum value of the integral compensation amount.

[0077] As an example, in step S404, the vehicle-mounted control device determines whether the initial compensation amount is between the integral compensation upper limit and the integral compensation lower limit, and determines the initial compensation amount to meet the preset requirement on the feedback compensation amount when the initial compensation amount is between the integral compensation upper limit and the integral compensation lower limit. is determined as the integral compensation amount. is not between the integral compensation upper limit and the integral compensation lower limit, the vehicle-mounted control device further determines whether the initial compensation amount is greater than the integral compensation upper limit, or the initial compensation amount is less than the integral compensation lower limit, and determines the initial compensation amount to meet the preset requirement on the feedback compensation amount when the initial compensation amount is greater than the integral compensation upper limit. is determined as the integral compensation amount. is less than the integral compensation lower limit. It can be understood that, in order to ensure the accuracy of compensation, the integral compensation amount is required to be between the integral compensation upper limit and the integral compensation lower limit, so as to avoid insufficient compensation, excessive error, and insufficient control accuracy, and to avoid excessive compensation, overshoot, and oscillation, and to affect the smoothness of vehicle control.

[0078] In this embodiment, the difference between the target speed and the current speed is integrated to determine a target integral value, and the integral compensation amount is determined according to the target integral value, which can effectively determine the difference between the current speed and the target speed, and determine a relatively accurate integral compensation amount. The initial compensation amount is processed according to the integral compensation upper limit and the integral compensation lower limit to determine the integral compensation amount, which can avoid insufficient compensation, excessive error, and insufficient control accuracy, and can avoid excessive compensation, overshoot, and oscillation, and can affect the smoothness of vehicle control.

[0079] In an embodiment, the comprehensive control instruction is the sum of the feedforward compensation amount, the feedback control amount, and the integral compensation amount.

[0080] As an example, the vehicle-mounted control device determines the sum of the feedforward compensation amount, the feedback control amount, and the integral compensation amount as the comprehensive control instruction. That is wherein, is a comprehensive control instruction, is a feedforward compensation amount, is a feedback control amount, is an integral compensation amount.

[0081] In this embodiment, the feedforward compensation amount, the feedback control amount and the integral compensation amount are comprehensively considered to obtain the comprehensive control instruction for controlling the vehicle to follow the preset track. In the case of guaranteeing the continuity and stability of the vehicle control according to the feedback control amount, the vehicle can not only be quickly and accurately controlled according to the feedforward compensation amount to improve the response speed of the controlled vehicle, but also can be error-compensated according to the integral compensation amount to improve the accuracy of the vehicle control, thereby improving the accuracy of the vehicle tracking the preset track.

[0082] The application also provides a vehicle-mounted control device 50, which comprises a processor 510 and a memory 520, wherein the memory 510 is used for storing a computer program, and the processor 520 is used for executing the program stored in the memory 510 to realize the vehicle control method introduced in any of the embodiments of the application. Figure 5

[0083] The application also provides a vehicle control system, which comprises the vehicle-mounted control device, the data acquisition device and the target component in the above embodiments, the vehicle-mounted control device is connected with the data acquisition device and the target component respectively, and is used for determining a comprehensive control instruction according to the current state data, the target state data and the current scene data of the vehicle acquired by the data acquisition device, and controlling the target component to work according to the comprehensive control instruction.

[0084] The application also provides a vehicle, which comprises the vehicle-mounted control device in the above embodiments.

[0085] The application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the vehicle control method introduced in any of the embodiments of the application.

[0086] In the application, multiple refers to two or more than two.

[0087] In the application, unless otherwise explicitly limited, the terms "mounting", "connection" and "connection" should be understood in a broad sense, for example, it can be fixed connection, or detachable connection, or integral connection; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or the communication between two elements. For those skilled in the art, the specific meaning of the above terms in the application can be understood according to the specific circumstances.

[0088] ​The terms "first", "second", "third", "fourth" and the like in the present application, if any, are used to distinguish similar objects, and do not necessarily indicate a particular order or sequence.

[0089] The term "and / or" in the present application is only used to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent three cases of A alone, A and B together, and B alone. In addition, the character " / " in the present application generally represents an "or" relationship between the associated objects before and after it.

[0090] If not specifically stated, all steps of the present application can be performed in sequence or randomly. For example, the method comprises steps A and B, which means that the method can comprise steps A and B performed in sequence, or steps B and A performed in sequence. For example, the method can further comprise step C, which means that step C can be added to the method in any order, for example, the method can comprise steps A, B and C, or steps A, C and B, or steps C, A and B, etc.

[0091] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modifications, equivalent replacements and improvements made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A vehicle control method characterized by, The method comprises: acquiring current state data corresponding to the vehicle, target state data, and current scene data, the current state data being state data collected at a current position, the target state data being preset state data corresponding to a target tracking point, and the current scene data being scene data in which the vehicle travels at the current position; determining a feedforward compensation amount based on the target state data; processing the current state data and the target state data using a state space model constructed based on a vehicle dynamics model to determine a feedback control amount; determining an integral compensation amount based on the current state data, the target state data, and the current scene data; determining a comprehensive control instruction according to the feedforward compensation amount, the feedback control amount, and the integral compensation amount, and controlling a target component to work according to the comprehensive control instruction.

2. The vehicle control method according to claim 1, characterized by, The target state data comprises a target acceleration corresponding to the target tracking point. The feedforward compensation amount is determined based on the target acceleration and a pre-labeled feedforward compensation parameter.

3. The vehicle control method according to claim 1, characterized by, The current state data comprises a current longitudinal speed and a current longitudinal position corresponding to the current position, and the target state data comprises a target longitudinal speed and a target longitudinal position corresponding to the target tracking point. The processing of the current state data and the target state data using the state space model constructed based on the vehicle dynamics model to determine the feedback control amount comprises: determining an objective function based on the current longitudinal speed and the current longitudinal position corresponding to the current position, and the target longitudinal speed and the target longitudinal position corresponding to the target tracking point; determining a constraint condition based on a control input amplitude limit and a control input rate limit in the state space model; performing a rolling solution on the state space model based on the constraint condition, and determining a target control sequence when the objective function is minimized, the target control sequence being a sequence formed by control inputs of a plurality of preset tracking points in a preset trajectory corresponding to a future time period, the target tracking point being the first preset tracking point in the preset trajectory; determining the feedback control amount corresponding to the target tracking point according to the target control sequence.

4. The vehicle control method according to claim 1, characterized by The determination of the integral compensation amount based on the current state data, the target state data, and the current scene data comprises: when at least one of the target state data and the current scene data satisfies a dynamic scene condition, determining the integral compensation amount according to the current state data and the target state data; when neither the target state data nor the current scene data satisfies the dynamic scene condition, determining the integral compensation amount as 0.

5. The vehicle control method according to claim 4, characterized by The current scene data comprises a planned behavior signal, and the target state data comprises a target acceleration and a target jerk corresponding to the target tracking point. The dynamic scene condition comprises a planned behavior enabling condition and a target state enabling condition. The planned behavior enabling condition comprises that the planned behavior signal is an emergency signal, and the emergency signal comprises an emergency acceleration signal and an emergency deceleration signal. The target state enabling condition comprises that an absolute value of the target acceleration is greater than a preset acceleration threshold, and an absolute value of the target jerk is greater than a preset jerk threshold.

6. The vehicle control method according to claim 4, characterized by The current state data includes a current speed corresponding to the current position; The target state data includes a target speed corresponding to the target tracking point; The determining the integral compensation amount according to the current state data and the target state data comprises: When at least one of the target state data and the current scene data satisfies a dynamic scene condition, determining an enabling signal value; Integrating a speed difference value corresponding to a target time period to determine a target integral value; the speed difference value is a difference between the target speed and the current speed; the target time period is a time interval corresponding to the current position and the target tracking point on a preset trajectory; Using the enabling signal and an integral gain coefficient to correct the target integral value to determine an initial compensation amount; According to an integral compensation upper limit and an integral compensation lower limit, processing the initial compensation amount to determine the integral compensation amount.

7. The vehicle control method according to claim 1, characterized by The comprehensive control instruction is a sum of the feedforward compensation amount, the feedback control amount and the integral compensation amount.

8. An in-vehicle control apparatus characterized by comprising: The vehicle control device comprises a processor and a memory, wherein, The memory is used for storing a computer program; The processor is used for executing the program stored on the memory to realize the vehicle control method in any one of claims 1-7.

9. A vehicle characterized by comprising: The vehicle control device comprises the vehicle control device in claim 8.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the vehicle control method in any one of claims 1-7. The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the vehicle control method in any one of claims 1-7.

Images