Vehicle control systems, methods, electronic devices, and media

Abstract

This application provides a vehicle control system, method, electronic device, and medium, relating to the field of vehicle technology. The vehicle control system includes a lateral control module and a longitudinal control module; wherein the lateral control module includes a first wheel steering angle control module, and the longitudinal control module includes a speed tracking control module and an acceleration compensation module. The first wheel steering angle control module obtains the first wheel steering angle control quantity of the first vehicle based on a path tracking control problem, where the first wheel is either the front wheel or the rear wheel; the speed tracking control module obtains the desired acceleration of the first vehicle based on a speed tracking control problem; the acceleration compensation module obtains the acceleration control quantity of the first vehicle based on the actual acceleration and / or vehicle driving resistance of the first vehicle, and based on the desired acceleration of the first vehicle. This application can balance the control performance and robustness of vehicle control.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a vehicle control system, method, electronic device and medium. Background Technology

[0002] Control performance and robustness are both important factors affecting the effectiveness of vehicle control. For example, the unknown longitudinal force induced by the vehicle's wheel steering angle affects the vehicle's longitudinal control performance. Therefore, it is necessary to take both control performance and robustness into account when designing vehicle control systems. Summary of the Invention

[0003] This application provides a vehicle control system, method, electronic device, and medium that can balance the control performance and robustness of vehicle control.

[0004] In a first aspect, embodiments of this application provide a vehicle control system, including: a vehicle lateral control module and a vehicle longitudinal control module; wherein, the vehicle lateral control module includes a first wheel steering angle control module, and the vehicle longitudinal control module includes a speed tracking control module and an acceleration compensation module; the first wheel steering angle control module is used to iteratively solve a nonlinear unconstrained optimization problem based on the minimum principle to obtain a first wheel steering angle control quantity of a first vehicle, wherein the first wheel is one of the front wheel and the rear wheel; wherein, the nonlinear unconstrained optimization problem is obtained by transforming a path tracking control problem for tracking a planned path using the obstacle function method and the Lagrange multiplier method, wherein the planned path includes a reference pose at a series of future time points, and the reference pose includes at least one of a reference lateral position and a reference heading angle; the speed tracking control module is used to solve a speed tracking control problem for tracking a planned speed to obtain a desired acceleration of the first vehicle; wherein, the planned speed includes a reference speed at a series of future time points; the acceleration compensation module is used to obtain an acceleration control quantity of the first vehicle based on at least one of the actual acceleration of the first vehicle and the vehicle driving resistance of the first vehicle, and based on the desired acceleration of the first vehicle.

[0005] In this embodiment, the first wheel steering angle control module and the speed tracking control module are independent of the vehicle chassis interface. They meet the control performance requirements through a designed high-performance control law to achieve precise control of path tracking and speed tracking. The acceleration compensation module, on the other hand, depends on the vehicle chassis interface and meets the requirements for resisting unknown interference through a designed robust control law to ensure the robustness of vehicle control. In this way, the vehicle control system can balance control performance and robustness.

[0006] This application embodiment decouples lateral control and longitudinal control through a hierarchical architecture, and decouples the two control quantities of front wheel steering angle and rear wheel steering angle in vehicle lateral control, and decouples control performance and robustness indicators in vehicle longitudinal control, thereby enabling vehicle control to exhibit high performance with "fast convergence, high precision and strong robustness".

[0007] Optionally, the vehicle lateral control module further includes a second wheel steering angle control module; the second wheel steering angle control module is used to obtain the second wheel steering angle control amount of the first vehicle based on the correlation between the front wheel steering angle and the rear wheel steering angle of the first vehicle and the first wheel steering angle control amount of the first vehicle, wherein the second wheel is the other of the front wheel and the rear wheel; wherein the correlation is obtained based on the maximum value of the front wheel steering angle and the maximum value of the rear wheel steering angle of the first vehicle.

[0008] The second wheel steering angle control module relies on the vehicle chassis interface and uses a robust control law to meet the requirements of resisting unknown interference, so as to ensure the robustness of the vehicle's lateral control.

[0009] Optionally, the path tracking control problem is established based on the pose error of the first vehicle relative to the Frenet coordinate system, with the origin of the Frenet coordinate system on the planned path. The path tracking control problem is used to solve for the first wheel steering angle control quantity that satisfies the lateral control constraints and minimizes the first function. The first function includes a path tracking error optimization term and a smooth control quantity optimization term. The path tracking error optimization term is obtained based on the pose error between the vehicle trajectory and the planned path, and the smooth control quantity optimization term is obtained based on the error of the first wheel steering angle of the vehicle at adjacent time points.

[0010] This allows the vehicle to accurately track the planned path and ensures the stability of the vehicle's lateral control.

[0011] Optionally, any solution result includes state variables reflecting pose error and control variables reflecting the first wheel steering angle at a series of future time points. The nonlinear unconstrained optimization problem is iteratively solved based on the minimum principle to obtain the first wheel steering angle control variable of the first vehicle. This includes: if a solution result from the previous iteration exists, the nonlinear unconstrained optimization problem is iteratively solved based on the previous iteration result to obtain the solution result for the current iteration; if both the inner and outer loop conditions are met, the first wheel steering angle control variable of the first vehicle is obtained based on the control variable obtained in the current iteration; wherein, the inner loop condition includes that the error between the state variables obtained in the previous and current iterations is not greater than a set error threshold, and the outer loop condition includes that the value of the first parameter in the nonlinear unconstrained optimization problem reaches a set number of thresholds; if the inner loop condition is not met, the steps for solving the nonlinear unconstrained optimization problem are executed again; if the inner loop condition is met and the outer loop condition is not met, the value of the first parameter is increased, and the steps for solving the nonlinear unconstrained optimization problem are executed again.

[0012] This helps ensure the accuracy of lateral vehicle control and supports the vehicle in tracking the planned path.

[0013] Optionally, the speed tracking control problem is derived from the longitudinal control model, which includes the relationship between the vehicle's desired acceleration, the vehicle's actual acceleration, and the parameters of the first-order inertial element. The speed tracking control problem is used to solve for the desired acceleration that satisfies the longitudinal control constraints and minimizes the second function. The second function includes the speed tracking error, which describes the error between the reference speed and the vehicle speed predicted based on the desired acceleration.

[0014] By establishing a speed tracking control problem based on the vehicle's longitudinal response delay characteristics and reducing the error between the reference speed and the predicted actual vehicle speed, accurate longitudinal tracking of the vehicle can be guaranteed.

[0015] Optionally, solving the speed tracking control problem for tracking the planned speed includes: iteratively constructing a third function for a series of future time points according to the speed tracking control problem, from back to front; obtaining the expected acceleration at each time point by minimizing the third function at each time point, where the expected acceleration of the first vehicle is the expected acceleration at the next time point; wherein, the third function at any time point is obtained based on the speed tracking error and expected acceleration at that time point, as well as the speed tracking error at the next time point; the speed tracking error at any time point is used to describe the error between the reference speed at that time point and the vehicle speed predicted based on the expected acceleration at the previous time point.

[0016] By using a reinforcement learning framework to iteratively solve the velocity tracking control problem to obtain the desired acceleration, accurate acquisition of the desired acceleration can be achieved.

[0017] Optionally, the vehicle driving resistance includes at least one of wheel steering angle resistance, slope resistance, and tire rolling resistance; the wheel steering angle resistance is obtained based on the first wheel steering angle control amount of the first vehicle, the equivalent longitudinal force of the first axle, and the equivalent lateral force of the first axle, wherein the equivalent longitudinal force of the first axle and the equivalent lateral force of the first axle are both obtained based on the first wheel steering angle control amount and the speed of the first vehicle; the slope resistance is obtained based on the weight of the first vehicle and the slope of the ground where the first vehicle is located; the tire rolling resistance is obtained based on the weight, the slope, and a set tire rolling resistance coefficient.

[0018] This allows for an accurate estimation of vehicle drag, which helps improve the accuracy of longitudinal vehicle tracking.

[0019] Secondly, embodiments of this application provide a vehicle control method, comprising: iteratively solving a nonlinear unconstrained optimization problem based on the minimum principle to obtain a first wheel steering angle control quantity for a first vehicle, wherein the first wheel is one of the front wheel and the rear wheel; wherein the nonlinear unconstrained optimization problem is obtained by transforming a path tracking control problem for tracking a planned path using the obstacle function method and the Lagrange multiplier method, wherein the planned path includes a reference pose at a series of future time points, and the reference pose includes at least one of a reference lateral position and a reference heading angle; solving a speed tracking control problem for tracking a planned speed to obtain a desired acceleration of the first vehicle; wherein the planned speed includes a reference speed at a series of future time points; and obtaining an acceleration control quantity for the first vehicle based on at least one of the actual acceleration of the first vehicle and the vehicle driving resistance of the first vehicle, and based on the desired acceleration of the first vehicle.

[0020] Thirdly, embodiments of this application provide an electronic chip, including: a processor for executing computer program instructions stored in a memory, wherein when the computer program instructions are executed by the processor, the electronic chip is triggered to perform the method as described in any of the first aspects.

[0021] Fourthly, embodiments of this application provide an electronic device including at least one processor and a memory coupled together. The memory is used to store computer program instructions, and the processor is used to execute the computer program instructions. When the computer program instructions are executed by the processor, the electronic device is triggered to perform a method as described in any of the first aspects.

[0022] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the method as described in any of the first aspects.

[0023] In a sixth aspect, embodiments of this application provide a computer program product, which includes a computer program that, when run on a computer, causes the computer to perform the method as described in any of the first aspects.

[0024] The technical effects of the aforementioned aspects can be referenced from each other, and will not be elaborated further here. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below.

[0026] Figure 1 A block diagram of a vehicle control system provided in this application embodiment;

[0027] Figure 2 A block diagram of another vehicle control system provided in this application embodiment;

[0028] Figure 3 A schematic diagram illustrating a vehicle kinematic model provided in an embodiment of this application;

[0029] Figure 4 A schematic diagram illustrating a path tracking error model provided in an embodiment of this application;

[0030] Figure 5 This is a schematic diagram illustrating the path tracking control problem provided in the embodiments of this application;

[0031] Figure 6 A schematic diagram illustrating the wheel steering angle resistance provided in an embodiment of this application;

[0032] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0033] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0034] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0035] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0036] It should be understood that the term "at least one" as used in this document refers to one or more, and "more than one" refers to two or more. The term "and / or" as used in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. A and B can be singular or plural. Additionally, the character " / " in this document generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0037] It should be understood that although the terms "first," "second," etc., may be used to describe the set thresholds in the embodiments of this application, these set thresholds should not be limited to these terms. These terms are only used to distinguish the set thresholds from each other. For example, without departing from the scope of the embodiments of this application, the first set threshold may also be referred to as the second set threshold, and similarly, the second set threshold may also be referred to as the first set threshold.

[0038] refer to Figure 1 One embodiment of this application provides a vehicle control system for controlling a vehicle 100. This vehicle control system features a decoupled design for lateral and longitudinal control. The longitudinal control module includes an outer-loop longitudinal control module 101 and an inner-loop longitudinal control module 102. The lateral control module includes a top-level lateral control module 103 and a bottom-level lateral distribution module 104. The inner-loop longitudinal control module 102 includes a feedforward control module 1021 and a feedback control module 1022.

[0039] For example, vehicle 100 can be a four-wheel steering vehicle, a front-wheel steering vehicle, a rear-wheel steering vehicle, a steer-by-wire vehicle, or a non-steer-by-wire vehicle. Taking a front-wheel steering vehicle as an example, compared to a front-wheel steering vehicle (where the front wheels are the drive wheels), a four-wheel steering vehicle can simultaneously control the front and rear wheels to achieve steering actions by adding rear wheel steering angle control. This can improve the vehicle's maneuverability during parking and driving. For example, a four-wheel steering vehicle can more easily complete parking operations in narrow environments.

[0040] Figure 1The vehicle control system shown is designed based on a hierarchical architecture, which decouples the lateral and longitudinal control, the front and rear wheel control in the lateral control, and the control performance and robustness indicators in the longitudinal control.

[0041] Figure 1 In the vehicle control system shown, the top-level lateral control module 103 solves for the front wheel steering angle control quantity that enables path tracking control based on the planned path; the bottom-level lateral allocation module 104 solves for the rear wheel steering angle control quantity based on the front wheel steering angle control quantity output by the top-level lateral control module 103 and the correlation between the front and rear wheel steering angles.

[0042] The planned path can include the vehicle's reference lateral position y at a series of future time points. ref and reference heading angle Lateral control of a vehicle can be achieved by controlling the steering angles of its front and rear wheels.

[0043] In one implementation, the top-level lateral control module 103 can calculate and output the desired front wheel steering angle based on the planned path tracking error (such as lateral position error and heading angle error).

[0044] The top-level lateral control module 103 is independent of the vehicle chassis interface. It uses high-performance control laws to meet control performance requirements and ensure vehicle lateral control performance. The bottom-level lateral distribution module 104 depends on the vehicle chassis interface. It uses robust control laws to meet the requirements of resisting unknown interference and ensure the robustness of vehicle lateral control.

[0045] Figure 1 In the vehicle control system shown, the outer ring longitudinal control module 101 calculates the desired acceleration 'a' to achieve speed tracking control based on the planned speed. c The inner ring longitudinal control module 102 compensates for acceleration error and driving resistance interference based on the desired acceleration output by the outer ring longitudinal control module 101, the actual vehicle acceleration (which can be obtained from the vehicle speed collected by the vehicle sensor), and the vehicle driving resistance, and obtains the acceleration control quantity.

[0046] The feedback control module 1022 can obtain the acceleration error based on the desired acceleration and the actual acceleration, and use the acceleration error as the disturbance feedback quantity; the feedforward control module 1021 can estimate the vehicle's driving resistance in real time, and use the estimated vehicle driving resistance as the disturbance feedforward quantity. Based on the disturbance feedback quantity and the disturbance feedforward quantity, disturbance compensation processing can be performed on the desired acceleration to obtain an acceleration control quantity that can cope with the disturbance.

[0047] The planned speed can include the vehicle's reference speed v at a series of future time points. refBy controlling the vehicle's acceleration, longitudinal control of the vehicle can be achieved.

[0048] The outer ring longitudinal control module 101 is independent of the vehicle chassis interface and meets the control performance requirements by designing a high-performance control law to ensure the longitudinal control performance of the vehicle. The inner ring longitudinal control module 102 depends on the vehicle chassis interface and meets the requirements for resisting unknown interference by designing a robust control law to ensure the robustness of the longitudinal control of the vehicle.

[0049] The vehicle control system can achieve vehicle driving control by decoupling the lateral and longitudinal directions of the vehicle.

[0050] in this way, Figure 1 The vehicle control system shown can have at least the following characteristics:

[0051] (1) The outer ring longitudinal control module and the top-level lateral control module are independent of the vehicle chassis interface. They meet the control performance requirements through the design of high-performance control laws to achieve precise control of path tracking and speed tracking. The inner ring longitudinal control module and the bottom-level lateral distribution module depend on the vehicle chassis interface. They meet the requirements of resisting unknown interference through the design of robust control laws to ensure the robustness of vehicle control. In this way, the vehicle control system can take into account both control performance and robustness.

[0052] (2) By decoupling lateral control and longitudinal control through a hierarchical architecture, and decoupling the two control quantities of front wheel steering angle and rear wheel steering angle in vehicle lateral control, and decoupling control performance and robustness indicators in vehicle longitudinal control, the vehicle control exhibits high performance with "fast convergence, high precision and strong robustness".

[0053] (3) Through the decoupling design of the lateral and longitudinal directions, the influence of the unknown longitudinal force induced by the two control variables of the front wheel steering angle and the rear wheel steering angle on the longitudinal control performance of the vehicle can be effectively suppressed, so that the longitudinal control of the vehicle can take into account both control performance and robustness requirements.

[0054] (4) Through the layered architecture design, the vehicle control requirements of "easy to implement, easy to maintain, easy to expand and easy to upgrade" are met.

[0055] (5) The vehicle control system has backward compatibility, meaning it can be applied not only to the vehicle control of four-wheel steering vehicles but also to the vehicle control of front-wheel steering vehicles. Therefore, it can serve as the basis for building a "high-performance, platform-based, and vehicle-model-decoupled" general vehicle control architecture.

[0056] (6) Figure 1When the vehicle control system shown is applied to the parking scenario of a four-wheel steering vehicle, the four-wheel steering vehicle can synchronously control the front and rear wheels to achieve steering actions based on the vehicle control system, and can accurately track and control the parking planning path, so that the four-wheel steering vehicle has better maneuverability during the parking process and can more easily complete the parking operation in narrow spaces.

[0057] refer to Figure 2 This application provides a vehicle control system 200, which includes a vehicle lateral control module 210 and a vehicle longitudinal control module 220. The vehicle lateral control module 210 includes a first wheel steering angle control module 211, and the vehicle longitudinal control module 220 includes a speed tracking control module 221 and an acceleration compensation module 222.

[0058] Feasibly, the vehicle control system 200 is applicable to vehicle control scenarios such as parking and driving. During parking, the vehicle is in a low-speed, large-angle movement state, and the vehicle control system 200 supports accurate and stable parking in parking scenarios.

[0059] Feasibly, the vehicle control system 200 is applicable to vehicles such as front-wheel steering vehicles, rear-wheel steering vehicles, and four-wheel steering vehicles. Specifically, when the vehicle control system 200 is used to control a front-wheel steering vehicle, the first wheel is the front wheel; when the vehicle control system 200 is used to control a rear-wheel steering vehicle, the first wheel is the rear wheel. When the vehicle control system 200 is used to control a four-wheel steering vehicle, the first wheel can be the front wheel.

[0060] In the vehicle control system 200, the first wheel steering angle control module 211 solves the nonlinear unconstrained optimization problem iteratively based on the minimum principle to obtain the first wheel steering angle control quantity of the first vehicle (the controlled vehicle). The first wheel is one of the front wheel and the rear wheel.

[0061] The nonlinear unconstrained optimization problem is obtained by transforming the path tracking control problem for tracking the planned path using the obstacle function method and the Lagrange multiplier method. The planned path includes the reference pose at a series of future time points, and the reference pose includes at least one of the reference lateral position and the reference heading angle.

[0062] The parking path tracking control problem solver constructed by integrating the obstacle function method, the Lagrange multiplier method, and the minimum principle can achieve high computational efficiency, thus improving the accuracy of vehicle lateral control by reducing the computation cycle of the top-level lateral control.

[0063] In one embodiment, the first wheel steering angle control module 211 may be the aforementioned top-level lateral control module 103.

[0064] Feasibly, the path tracking control problem is used to track a planned path by controlling the steering angle of the first wheel of a first vehicle. The steering angle of the first wheel of the first vehicle can be controlled based on the first wheel steering angle control value obtained from solving the path tracking control problem.

[0065] In one embodiment, the constraints of the path tracking control problem can be obtained from the vehicle control dynamics model, the maximum / minimum values ​​of the wheel steering angle, etc.

[0066] Taking the first wheel as the front wheel as an example, in one embodiment, based on the nonlinear unconstrained optimization problem obtained by transforming the path tracking control problem, the nonlinear unconstrained optimization problem can be solved repeatedly to obtain the front wheel steering angle control quantity that meets the path tracking requirements. Here, the path tracking requirement can be that the difference between the predicted trajectory and the planned path meets the requirements (i.e., controlling the vehicle to travel along the planned path laterally), and the predicted trajectory is the vehicle's predicted trajectory over a future period when the vehicle is controlled using the solved front wheel steering angle control quantity.

[0067] If the front wheel steering angle obtained in the current iteration does not meet the path tracking requirements, the next iteration of the nonlinear unconstrained optimization problem can be performed based on the front wheel steering angle obtained in the current iteration. This ensures that as the number of iterations increases, the obtained front wheel steering angle control value will be more likely to meet the path tracking requirements. In this way, the nonlinear unconstrained optimization problem can be solved repeatedly until a front wheel steering angle that meets the path tracking requirements is obtained. This front wheel steering angle is then used as the front wheel steering angle control value to control the vehicle's front wheel steering angle for lateral control.

[0068] The reference pose in the planned path can include a reference lateral position and / or a reference heading angle. By using the reference lateral position and / or reference heading angle as a reference to solve the front wheel steering angle, and using the solved front wheel steering angle to control the vehicle, the vehicle can achieve the effect of lateral control that tracks the planned path.

[0069] The planned path can include reference poses at a series of future time points, such as the reference poses at steps 0 to N. Based on the vehicle's current state variables (including pose) and control variables (including front wheel steering angle), the vehicle's state variables at the next time step can be predicted. Thus, in any given iteration, the state variables and control variables at steps 0 to N can be predicted sequentially, and the resulting series of state variables reflects the predicted trajectory.

[0070] In the vehicle control system 200, the speed tracking control module 221 solves the speed tracking control problem for tracking the planned speed to obtain the expected acceleration of the first vehicle; wherein, the planned speed includes the reference speed at a series of future time points.

[0071] In one embodiment, the speed tracking control module 221 may be the aforementioned outer ring longitudinal control module 101.

[0072] Feasibly, the speed tracking control problem can be used to track a planned speed by controlling the acceleration of a first vehicle. The acceleration of the first vehicle can be controlled based on the desired acceleration obtained from solving the speed tracking control problem, after acceleration disturbance compensation.

[0073] In one embodiment, the constraints of the speed tracking control problem can be obtained from a vehicle longitudinal control model, etc.

[0074] In the vehicle control system 200, the acceleration compensation module 222 obtains the acceleration control amount of the first vehicle based on at least one of the actual acceleration of the first vehicle and the vehicle driving resistance of the first vehicle, as well as based on the desired acceleration of the first vehicle.

[0075] The vehicle longitudinal control module can control the acceleration of the first vehicle based on the obtained acceleration control value.

[0076] In one embodiment, the acceleration compensation module 222 may be the aforementioned inner ring longitudinal control module 102.

[0077] The acceleration compensation module 222 can estimate the vehicle's driving resistance as a disturbance factor in real time through the feedforward control module 1021. By combining the vehicle's driving resistance with the acceleration control value, the impact of the vehicle's driving resistance on the vehicle's longitudinal control robustness can be optimized.

[0078] The acceleration compensation module 222 can obtain the acceleration error, a disturbance factor, based on the vehicle's reference acceleration and actual acceleration through the feedback control module 1022. By combining the acceleration error to obtain the acceleration control quantity, the impact of the acceleration error on the vehicle's longitudinal control robustness can be optimized.

[0079] By compensating for acceleration errors and vehicle drag, an acceleration control quantity that meets the speed tracking requirements can be obtained. The speed tracking requirement is defined as the difference between the predicted speed and the planned speed meeting certain conditions (i.e., controlling the vehicle longitudinally at the planned speed). The predicted speed is the vehicle's predicted speed over a future period when controlled by the obtained acceleration control quantity.

[0080] Since the acceleration control quantity obtained after compensation eliminates the influence of interference factors on the longitudinal control of the vehicle, the acceleration control quantity can better meet the speed tracking requirements compared with the reference acceleration, thus ensuring the longitudinal control effect of the vehicle.

[0081] To ensure that the vehicle control system 200 is applicable not only to four-wheel steering vehicles but also backward compatible to front-wheel or rear-wheel steering vehicles, in one embodiment of this application, reference is made to... Figure 2 The vehicle lateral control module 210 also includes a second wheel steering angle control module 212; the second wheel steering angle control module 212 obtains the second wheel steering angle control amount of the first vehicle based on the correlation between the front wheel steering angle and the rear wheel steering angle of the first vehicle and the first wheel steering angle control amount of the first vehicle, wherein the second wheel is the other of the front wheel and the rear wheel; wherein the correlation is obtained based on the maximum value of the front wheel steering angle and the maximum value of the rear wheel steering angle of the first vehicle.

[0082] The steering angle of the second wheel of the first vehicle can be controlled based on the obtained second wheel steering angle control value.

[0083] In one embodiment, the second wheel steering angle control module 212 can be the aforementioned underlying lateral allocation module 104.

[0084] After the top-level lateral control module obtains the front wheel steering angle control value, the bottom-level lateral distribution module can calculate and output the rear wheel steering angle control value accordingly, thereby achieving perfect decoupled control of the front wheel steering angle and the rear wheel steering angle of a four-wheel steering vehicle.

[0085] The second wheel steering angle control module relies on the vehicle chassis interface and uses a robust control law to meet the requirements of resisting unknown interference, so as to ensure the robustness of the vehicle's lateral control.

[0086] Taking front-wheel steering vehicles as an example, compared to front-wheel steering vehicles, the additional rear-wheel steering angle control of four-wheel steering vehicles causes the instantaneous steering center of the vehicle to change in real time, making it difficult to select a vehicle reference point for lateral control. However, this embodiment of the application decouples parking lateral control into a top-level lateral control module and a bottom-level lateral distribution module, achieving perfect decoupling of the two control quantities—the front-wheel steering angle and the rear-wheel steering angle—in four-wheel steering vehicles, and fixing the vehicle reference point.

[0087] The allocation strategy for the front and rear wheel steering angles, two control variables in four-wheel steering vehicles, is a crucial factor affecting the vehicle's lateral control performance. However, the complex nonlinear relationship between these two control variables and the lateral motion of the four-wheel steering vehicle presents a significant challenge to formulating the allocation strategy. This application, by decoupling the front and rear wheel steering angles and combining the correlation between them, achieves a reasonable allocation of the front and rear wheel steering angle control variables, thereby improving the accuracy of lateral control.

[0088] refer to Figure 3The schematic diagram shown reflects the vehicle kinematics model of the four-wheel steering vehicle 301. In order to make the closed-loop control effect of the front and rear wheel steering angles of the four-wheel steering vehicle more human-like, safe and comfortable, in one embodiment, the rear wheel steering angle δ can be defined. r With front wheel steering angle δ f The action is performed and satisfies the following equation (1).

[0089] L f tanδ r =L r tanδ f (1)

[0090] In equation (1), L f and L r These are the distances from point P on a four-wheel steering vehicle to the front and rear axles, respectively.

[0091] In equation (1),

[0092] Among them, the maximum steering angles of the front and rear wheels of a four-wheel steering vehicle are δ fmax and δ rmax L is the wheelbase of a four-wheel steering vehicle.

[0093] As can be seen from equation (1), the bottom-level lateral distribution module can calculate the output rear wheel steering angle control quantity based on the front wheel steering angle control quantity output by the top-level lateral control module by comprehensively considering the maximum value constraint of the front and rear wheel steering angles of the four-wheel steering vehicle.

[0094] For example, the relationship between the front and rear wheel steering angles of the first vehicle can be expressed as Equation (1). Then, by substituting the first wheel steering angle control quantity output by the first wheel steering angle control module 211 into Equation (1), the second wheel steering angle control quantity of the first vehicle can be obtained.

[0095] In one embodiment of this application, the path tracking control problem is established based on the pose error of the first vehicle relative to the Frenet coordinate system, the origin of which is on the planned path. The path tracking control problem is used to solve for the first wheel steering angle control quantity that satisfies the lateral control constraint and minimizes the first function. The first function includes a path tracking error optimization term and a smooth control quantity optimization term. The path tracking error optimization term is obtained based on the pose error between the vehicle trajectory and the planned path, and the smooth control quantity optimization term is obtained based on the error of the first wheel steering angle of the vehicle at adjacent time points.

[0096] The path tracking control problem is used to solve for the first wheel steering angle control value. Therefore, a path tracking control problem including the first wheel steering angle control value can be established based on the Frenet coordinate system. Thus, the lateral and longitudinal directions mentioned in this embodiment can be the lateral and longitudinal directions based on the Frenet coordinate system.

[0097] Pose error can include the error between the vehicle's actual lateral position and the reference lateral position, as well as the error between the vehicle's actual heading angle and the reference heading angle. In the process of iteratively solving the nonlinear unconstrained optimization problem, the vehicle's actual lateral position at a future time can be a prediction of the vehicle's lateral position, and the vehicle's actual heading angle at a future time can be a prediction of the vehicle's heading angle.

[0098] In one embodiment, the lateral control constraints can be obtained from the vehicle control dynamics model, the maximum / minimum values ​​of the wheel steering angle, etc.

[0099] By including a path tracking error optimization term in the first function and solving for the wheel steering angle control quantity that minimizes the first function, the error between the vehicle trajectory and the planned path can be reduced. Furthermore, by including a smoothing control optimization term in the first function and solving for the wheel steering angle control quantity that minimizes the first function, the stability of the vehicle's lateral control can be guaranteed.

[0100] In one embodiment of this application, any solution result includes a state variable reflecting the pose error at a series of future time points and a control variable reflecting the first wheel steering angle. Based on this, the step of iteratively solving the nonlinear unconstrained optimization problem based on the minimum principle to obtain the first wheel steering angle control variable of the first vehicle may include:

[0101] If the solution result of the previous iteration exists, the nonlinear unconstrained optimization problem is iteratively solved based on the solution result of the previous iteration to obtain the solution result of the current iteration. If both the inner and outer loop conditions are met, the first wheel steering angle control quantity of the first vehicle is obtained based on the control quantity obtained in the current iteration. The inner loop condition includes that the error between the state quantities obtained in the previous and current iterations is not greater than a set error threshold, and the outer loop condition includes that the value of the first parameter in the nonlinear unconstrained optimization problem reaches a set number of thresholds.

[0102] Since the solution process in the first loop does not have the solution result from the previous loop, the first loop solution process can be performed based on the preset initial information. Subsequent loop solutions are all based on the solution result of the previous loop (i.e., the (k+1)th loop solution is performed based on the solution result of the kth loop).

[0103] The iterative solution of nonlinear unconstrained optimization problems can be achieved by combining inner and outer loops.

[0104] In any given outer loop solution process, the inner loop solution process can be executed multiple times until the inner loop solution result meets the inner loop condition, at which point the current outer loop solution process ends, and it is determined whether to execute the next outer loop solution process. After executing the set number of outer loop solution processes, the first wheel steering angle control value of the first vehicle is obtained from the final solution result (i.e., the solution result of the last inner loop process in the last outer loop solution process).

[0105] In one embodiment, the error threshold can be set to a minimum value, such as 10. -3 The inner loop is stopped when the difference between the solutions of two adjacent inner loops is extremely small. At this point, the solution result reaches the inner loop optimum in the current outer loop solution process.

[0106] In one embodiment of this application, if the inner loop condition is not met, the step of solving the nonlinear unconstrained optimization problem is executed again, that is, the inner loop solution process is executed again. Each inner loop solution process is based on the solution result of its previous loop, so that as the number of inner loops increases, the solution result of the inner loop gradually tends to the optimal solution result of the current outer loop solution process.

[0107] In one embodiment of this application, if the inner loop condition is met and the outer loop condition is not met, the value of the first parameter is increased, and the step of solving the nonlinear unconstrained optimization problem is executed again, that is, the outer loop solution process is executed again.

[0108] Since the solution obtained through multiple iterations of the inner loop in each outer loop iteration reaches the optimal result of the current outer loop iteration, changing the value of the first parameter can escape the local optimum constraint, thus ensuring the effectiveness of the next outer loop iteration. This process, through multiple outer loop iterations, leads to a globally optimal solution. This helps ensure the accuracy of lateral vehicle control and supports the vehicle's ability to track and plan its path.

[0109] The following describes a feasible implementation process for the first wheel steering angle control module 211 to solve the front wheel steering angle control quantity, taking the first wheel as the front wheel, the vehicle as a four-wheel steering car, and the planned path as a planned parking path as an example. The related technical implementation is also applicable to embodiments of vehicles with the first wheel as the rear wheel, vehicles with front or rear wheel steering, and planned paths as planned driving paths.

[0110] refer to Figure 3 Point P on a four-wheel steering vehicle can be used as a reference point to model path tracking error. For example... Figure 4 As shown, the lateral (or sideways) error E of point P on the four-wheel steering vehicle 401 relative to the projected point on the parking path 402 can be defined. y and heading angle error E ψ Equations (2) and (3) are respectively.

[0111]

[0112] E ψ =ψ-ψ σ (3)

[0113] In equations (2) and (3), (X, Y, ψ) represents the pose of point P on a four-wheel steering vehicle in the fixed coordinate system OXY, (X, Y, ψ) represents the pose of point P on the vehicle. σ Y σ , ψ σ (x) represents the pose of the projection point on the parking path in the fixed coordinate system OXY.

[0114] The lateral velocity of the vehicle can be assumed to be v. y If the value is zero, then the lateral error E of point P on a four-wheel steering vehicle relative to the projected point on the parking path is zero. y (or lateral position error) and heading angle error E ψ Differentiating, we can obtain the following equations (4) and (5).

[0115]

[0116]

[0117] In equation (4), v x This refers to the longitudinal speed of a four-wheel steering vehicle.

[0118] From the instantaneous center of velocity theorem, we can obtain the following equation (6).

[0119]

[0120] In the formula, ρ σ Let be the radius of curvature of the parking path at the projection point. v is the velocity of the projected point on the parking path generated by the movement of a four-wheel steering vehicle along the parking path. σ The longitudinal speed v of a four-wheel steering vehicle x Projected velocity along the tangent of the parking path, radius of curvature ρ σ Characterizing the inherent geometric properties of the parking path, the longitudinal velocity v of a four-wheel steering vehicle. x This characterizes the actual motion state of a four-wheel steering vehicle along the parking path. From equation (6), it can be seen that, based on the instantaneous center of velocity theorem, the instantaneous rate of change of the heading angle of the parking path at the projection point can be derived.

[0121] Simplifying equation (6), we obtain the velocity of the projected point on the parking path generated by the movement of a four-wheel steering vehicle along the parking path. The expression for is given by equation (7).

[0122]

[0123] In equation (7), k s Let k be the curvature of the parking path at the projection point. s =1 / ρ σ .

[0124] Define the state vector as ζ = [E y E ψ ] T By the chain rule, we can obtain equation (8).

[0125]

[0126] Based on the chain derivative of equation (8), combined with equations (4), (5), and (7), we can obtain equations (9) and (10).

[0127]

[0128]

[0129] In equation (10), k v The curvature of the actual driving path of a four-wheel steering vehicle can be expressed as:

[0130] Based on the inherent properties of the steering mechanism of a four-wheel steering vehicle, the curvature k of the actual driving path of the four-wheel steering vehicle is known. v satisfy:

[0131] Define the parameterized state vector as ζ(s)=[E y (s) E ψ (s)] T The control quantity is k v Based on the above derivation, the parameterized state equation can be obtained as follows (11).

[0132]

[0133] From equation (11), it can be seen that the lateral error E of point P on a four-wheel steering vehicle relative to the projected point on the parking path is... y and heading angle error E ψ After parameterization, their dynamic characteristics are independent of the longitudinal speed v of the four-wheel steering vehicle. xIt has dynamic characteristics and allows control of the system to be applied even when the system is stopped.

[0134] like Figure 5 As shown, the Euclidean distance Δs between adjacent projection points on parking path 502 and the curvature k at the projection point are used. s Characterizing the geometric properties of the parking path, and establishing a series of Frenet coordinate systems with the projection points on the parking path as the origin, it can be seen that the parameterized state equation of equation (11) describes the dynamic pose information of the four-wheel steering vehicle relative to the Frenet coordinate system. Therefore, the parking path tracking control problem can be established as equation (12).

[0135] minJ error +J smooth +ηα T a

[0136]

[0137] In equation (12), J error and J smooth Let A and B be the parking path tracking error optimization term and the smoothing control quantity optimization term, respectively. A and B are both discretization coefficient matrices of the parameterized state equation, and and represent the relaxation vector and weight coefficients, respectively. α and η are the upper boundaries of the relaxation vector elements, respectively. and These are the lower and upper boundaries of the lateral error of point P on a four-wheel steering vehicle relative to the projected point on the parking path, E. ymin and E ψmin These are the lower and upper boundaries of the heading angle error of point P on a four-wheel steering vehicle relative to its projection point on the parking path, s k This is the path length at step k.

[0138] Among them, J error and J smooth A, B, and α can be represented as:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144] Among them, γ1, γ2 and γ3 are all weighting coefficients.

[0145] In one embodiment, the path tracking control problem can be the parking path tracking control problem of equation (12), then the first function mentioned above can be J in equation (12). error +J smooth +ηα T a, The first function includes the path tracking error optimization term J error And smooth control quantity optimization term J smooth Path tracking error optimization term J error Based on the pose error (lateral error E) between the vehicle trajectory and the planned path y and heading angle error E ψ The smooth control quantity optimization term J is obtained. smooth Based on the vehicle's first wheel steering angle at adjacent moments The error was obtained.

[0146] Thus, by solving for the first wheel steering angle control quantity that satisfies the lateral control constraints (i.e., the equality and inequality constraints in equation (12) above) and minimizes the first function, the path tracking error optimization term J can be optimized. error Smaller, smoother control quantity optimization term J smooth Smaller. Based on the smaller smooth control quantity optimization term J smooth This can reduce the error between the vehicle trajectory and the planned path, supporting the vehicle to accurately track the planned path while driving, and is based on a smaller smooth control quantity optimization term J. smooth This ensures the stability of the vehicle's lateral control.

[0147] After obtaining the parking path tracking control problem in equation (12) above, the lateral error and heading angle error of point P on the four-wheel steering vehicle relative to the projection point on the parking path can be obtained by solving equation (12). Furthermore, the pose of point P on the four-wheel steering vehicle in the fixed coordinate system OXY can be obtained by coordinate transformation (see equation (13) below).

[0148]

[0149] Let x k =[E y (s k E ψ (s k )] T u k =k v (s k Then, the parking path tracking control problem described by equation (12) can be abstracted into the following nonlinear constrained optimization problem (see equation (14) below).

[0150]

[0151] See equations (15) and (16) below, and adopt the inequality constraints in the following obstacle function expression (14).

[0152] b k (x k ) = -log(-g k (x k k = 0, 1, ..., N (15)

[0153] d k (u k ) = -log(-h k (u k k = 0, 1, ..., N-1 (16)

[0154] When the inequalities in equation (14) satisfy the constraints, the function values ​​of equations (15) and (16) are 0; when the inequalities in equation (14) do not satisfy the constraints, the function values ​​of equations (15) and (16) are positive infinity. Therefore, by using equations (15) and (16), the nonlinear constrained optimization problem described by equation (14) can be transformed into the nonlinear constrained optimization problem described by equation (17).

[0155]

[0156] In equation (17), t is a parameter whose value gradually increases with the number of outer loop cycles.

[0157] Furthermore, the nonlinear constrained optimization problem described by equation (17) can be simplified to equation (18).

[0158]

[0159] In equation (18), φ N (x N )=tc N (x N )+b N (x N ),

[0160] The nonlinear constrained optimization problem described by equation (18) can be transformed into a nonlinear unconstrained optimization problem described by equation (19) using the Lagrange multiplier method.

[0161]

[0162] In equation (19), λ k+1 It is a Lagrange multiplier.

[0163] based on and For φ in equation (19) N (xN L k (x k ) and M k (u k )exist Performing a second-order Taylor expansion within the domain yields the following equations (20) to (22).

[0164]

[0165]

[0166]

[0167] in, This represents the value of x obtained in the previous iteration. N , This represents the value of x obtained in the previous iteration. k , This represents the result u obtained in the previous loop. k .

[0168] based on and For f in equation (19) k (x k ,u k )exist Performing a first-order Taylor expansion within the domain yields the following equation (23).

[0169]

[0170] Substituting equations (20) to (23) into equation (19), we obtain the unconstrained quadratic optimization problem described by equation (24).

[0171]

[0172] Equation (24) is an unconstrained quadratic optimization problem, and its local optimal solution is also the global optimal solution. Taking the partial derivative of equation (24), we can obtain the following equation (25).

[0173]

[0174] Furthermore, based on equation (25), equation (26) can be obtained.

[0175]

[0176] Assume λ k With δx k The following equation (27) applies between them.

[0177] λ k =P kδx k +υ k (27)

[0178] In equation (27), P k υ is the gain coefficient. k This is the deviation coefficient.

[0179] From the fourth sub-equation in equation (26), we can see that P N and υ N They can be expressed as equation (28) and equation (29) respectively.

[0180]

[0181]

[0182] Furthermore, according to equation (27), we can have equation (30).

[0183] λ k+1 =P k+1 δx k+1 +υ k+1 (30)

[0184] Substituting equations (27) and (30) into the first sub-equation of equation (26), we get equation (31).

[0185]

[0186] Further simplification of equation (31) yields equation (32).

[0187]

[0188] Substituting equation (30) into the second sub-equation of equation (26), we obtain equation (33).

[0189]

[0190] Substituting equation (33) into the third sub-equation in equation (26), we obtain equation (34).

[0191]

[0192] Substituting equation (32) into equation (34), we obtain equation (35).

[0193]

[0194] Equation (36) can be obtained from equation (35).

[0195]

[0196] Further simplification of equation (36) yields equations (37) and (38).

[0197]

[0198]

[0199] Based on the above, the solution steps for the parking path tracking control problem can include the following steps 1.1 to 1.7:

[0200] Step 1.1: Initialize solver parameters.

[0201] Step 1.2, based on the solution results of the previous loop. Constructing the nonlinear constrained optimization problem of formula (18).

[0202] Step 1.3: Using equations (28) and (29) as initial conditions, calculate the gain coefficient P iteratively from back to front (i.e. from step k+1 to step k) using equations (36) and (37). k And deviation coefficient υ k That is, calculate P1, P2, ..., P N and υ1,υ2,...,υ N .

[0203] Step 1.4: Calculate δu using equation (33) iteratively from front to back (i.e., from step k to step k+1). k and according to Calculate u in the current iteration. k .

[0204] See equation (33), gain coefficient P k And deviation coefficient υ k To calculate δu k The intermediate variables required.

[0205] Step 1.5, calculate δx using the third sub-equation in equation (26). k and according to Calculate x in the current iteration. k .

[0206] After calculating uk and xk for the current iteration, the solution result (xk) for the current iteration is obtained. k x N u k After that, the solution result (x) can be used as a basis for the current iteration. k x N u k )right Update the solution so that when step 1.2 is executed again, the result (x) is based on the solution of the current iteration.k x N u k The nonlinear constraint optimization problem of equation (18) is constructed, and then the next iteration is performed.

[0207] Step 1.6, if δx k Satisfying the inner loop stopping criterion (e.g., |δx) k |≤10 -3 If δx k If the inner loop stopping criterion is not met, jump to step 1.2 to perform another inner loop solution.

[0208] Step 1.7: If t (see equation (17)) satisfies the outer loop stopping criterion (e.g., t≥100), the solution ends; if t does not satisfy the outer loop stopping criterion, t is incremented by 1, and the process jumps to step 1.2 to perform another outer loop solution.

[0209] After the solution is completed, the front wheel steering angle control value can be obtained based on the latest solution results.

[0210] In one embodiment, any solution obtained by the first wheel steering angle control module 211 through iteratively solving the nonlinear unconstrained optimization problem includes a state variable x reflecting the pose error. k =[E y (s k E ψ (s k )] T and the control quantity u used to reflect the steering angle of the first wheel k =k v (s k The first parameter can be t in equation (17), the inner loop condition can be the inner loop stopping criterion, and the outer loop condition can be the outer loop stopping criterion.

[0211] In one embodiment of this application, the speed tracking control problem is obtained based on a longitudinal control model, which includes the desired acceleration of the vehicle, the actual acceleration of the vehicle, and the correlation between the parameters of the first-order inertial element. The speed tracking control problem is used to solve for the desired acceleration that satisfies the longitudinal control constraints and minimizes the second function. The second function includes a speed tracking error, which describes the error between the reference speed and the vehicle speed predicted based on the desired acceleration.

[0212] The longitudinal response characteristics of a four-wheel steering vehicle can be approximated using a first-order inertial element, and a longitudinal control model for the vehicle can be constructed accordingly. By establishing a speed tracking control problem based on the vehicle's longitudinal response delay characteristics, the desired acceleration can be accurately obtained.

[0213] In one embodiment, the longitudinal control constraints can be obtained from a vehicle longitudinal control model, etc.

[0214] By incorporating the speed tracking error into the second function and solving for the desired acceleration that minimizes the second function, the error between the reference speed and the predicted actual vehicle speed can be reduced, ensuring accurate longitudinal tracking of the vehicle. Specifically, during the iterative solution of the speed tracking control problem, the actual vehicle speed at future moments can be used as a prediction of the vehicle speed.

[0215] In one embodiment of this application, solving the speed tracking control problem for tracking the planned speed includes: according to the speed tracking control problem, iteratively constructing a third function at a series of future time points from back to front (i.e. from step k+1 to step k) in chronological order; obtaining the expected acceleration at each time point by minimizing the third function at each time point, wherein the expected acceleration of the first vehicle is the expected acceleration at the next time point obtained.

[0216] The third function at any given time point is obtained based on the speed tracking error and expected acceleration at that time point, as well as the speed tracking error at the next time point. The speed tracking error at any given time point is used to describe the error between the reference speed at that time point and the vehicle speed predicted based on the expected acceleration at the previous time point.

[0217] By incorporating the speed tracking error of the current and next time moments into the third function, and solving for the expected acceleration that minimizes the third function, the error between the reference speed and the predicted actual vehicle speed can be reduced. This also supports iterative construction of the third function from back to front, thereby iteratively calculating the expected acceleration at each time point.

[0218] The speed tracking control module 221 establishes a speed tracking control problem based on the longitudinal response delay characteristics of the vehicle, and uses a reinforcement learning framework to iteratively solve the speed tracking control problem to obtain the desired acceleration, thus achieving accurate acquisition of the desired acceleration.

[0219] The following example illustrates a feasible implementation process for the speed tracking control module 221 to solve for the desired acceleration, using a four-wheel steering vehicle and a planned parking path as examples. The related technical implementation is also applicable to embodiments where the vehicle is a front-wheel or rear-wheel steering vehicle and the planned driving path is a planned driving path.

[0220] If the longitudinal response characteristics of a four-wheel steering vehicle are approximated by a first-order inertial element, then the nominal model of the longitudinal control of the outer ring of the parking vehicle for a four-wheel steering vehicle can be expressed as Equation (38).

[0221]

[0222] In equation (38), v and a are the actual speed and actual acceleration of a four-wheel steering vehicle, respectively; a c Let τ be the desired acceleration of a four-wheel steering vehicle; τ is the parameter of the first-order inertial element.

[0223] Define the system state vector and control vector as x k =[v k ,a k ] T and u k =[a c ] T If the sampling period is T, then equation (38) can be discretized into equation (39).

[0224] x k+1 =A k x k +B k u k (39)

[0225] In equation (39), the coefficient matrix A k Represented as B k Represented as

[0226] The objective of the longitudinal control of the outer ring of a four-wheel steering vehicle for parking is to enable the vehicle speed to quickly, accurately and stably track the parking reference speed under the condition of satisfying the control constraints. Thus, a constraint optimization problem as shown in Equation (40) can be defined to describe the longitudinal control of the outer ring of a four-wheel steering vehicle for parking.

[0227]

[0228] For the constrained optimization problem described by equation (40), the value function for the Nth step is equation (41).

[0229]

[0230] The action-value function for step N-1 is given by equation (42).

[0231]

[0232] Using x N =A N-1 x N-1 +B N-1 u N-1 The action-value function of step N-1 can be transformed into equation (43).

[0233]

[0234] Minimize the action-value function described by equation (43) to obtain the control law for step N-1 as equation (44).

[0235]

[0236] Substituting equation (44) into equation (43), we obtain the value function for the (N-1)th step as V(x) N-1 Simultaneously, the action-value function Q(x) for the (N-2)th step is constructed using the patterns described by equations (41) and (42). N-2 ,u N-2 Minimize Q(x) N-2 ,u N-2 Obtain the control law u for step N-2. N-2 By continuing in this way, the control law, value function, and action-value function for each step can be obtained iteratively from back to front.

[0237] Based on the above, the solution steps for the longitudinal control problem of the outer ring of a four-wheel steering vehicle during parking can include the following steps 2.1 to 2.3:

[0238] Step 2.1, initialize the solver parameters.

[0239] Step 2.2: Iteratively calculate the control law u for each step from back to front. k .

[0240] Step 2.3: Use the control law calculated iteratively for the next moment as the calculation output of the longitudinal control of the outer loop of the parking system.

[0241] Among them, u k =[a c ] T Then, the desired acceleration can be obtained based on the control variable.

[0242] In one embodiment, the speed tracking control problem solved by the speed tracking control module 221 can be Equation (40), and the third function mentioned above can be Equation (42).

[0243] In one embodiment of this application, the vehicle's driving resistance includes at least one of wheel steering angle resistance, slope resistance, and tire rolling resistance. This allows for accurate estimation of the vehicle's driving resistance, thereby helping to improve the accuracy of vehicle longitudinal tracking.

[0244] In one embodiment, the acceleration compensation module 222 can be the inner ring longitudinal control module 102 mentioned above. The inner ring longitudinal control module 102 can estimate the wheel steering angle resistance, slope resistance and tire rolling resistance in real time online, and calculate the vehicle driving resistance in combination with formula (45).

[0245] F x =F δ+F θ +F r (45)

[0246] Among them, F x F δ F θ and F r These are vehicle driving resistance, wheel steering angle resistance, slope resistance, and tire rolling resistance, respectively.

[0247] Feasibly, the wheel steering angle resistance is obtained based on the first wheel steering angle control amount of the first vehicle, the equivalent longitudinal force of the first wheel axle, and the equivalent lateral force of the first wheel axle, wherein the equivalent longitudinal force of the first wheel axle and the equivalent lateral force of the first wheel axle are both obtained based on the first wheel steering angle control amount of the first vehicle and the speed of the first vehicle.

[0248] In one embodiment, reference Figure 6 The wheel steering angle resistance F can be calculated according to equation (46). δ .

[0249] F δ =F x1 cosδ f -F y1 sinδ f +F x2 cosδ r +F y2 sinδ r (46)

[0250] In equation (46), F x1 and F y1 F represents the equivalent longitudinal force and equivalent lateral force on the front axle of the vehicle. x2 and F y2 These are the equivalent longitudinal force and equivalent lateral force of the rear axle of the vehicle.

[0251] In one feasible implementation, the vehicle speed and wheel steering angle are used as inputs, and the equivalent longitudinal force and equivalent lateral force of the front / rear axle are obtained by looking up a table.

[0252] It is feasible to obtain the ramp resistance based on the weight of the first vehicle and the slope of the ground where the first vehicle is located.

[0253] In one embodiment, the ramp resistance F can be calculated according to equation (47). θ .

[0254] F θ =mgsinθ (47)

[0255] In equation (47), m is the mass of the vehicle, g is the gravitational acceleration, and θ is the slope of the ground where the vehicle is located.

[0256] It is feasible to obtain the tire rolling resistance based on gravity, slope and a set tire rolling resistance coefficient.

[0257] In one embodiment, the tire rolling resistance F can be calculated according to equation (48). r .

[0258] F r =ξmgcosθ (48)

[0259] In equation (48), ξ is the tire rolling resistance coefficient.

[0260] This application provides a vehicle control method, which may include the following steps: Iteratively solving a nonlinear unconstrained optimization problem based on the minimum principle to obtain a first wheel steering angle control quantity for a first vehicle, wherein the first wheel is one of the front wheel and the rear wheel; wherein the nonlinear unconstrained optimization problem is obtained by transforming a path tracking control problem for tracking a planned path using the obstacle function method and the Lagrange multiplier method, wherein the planned path includes a reference pose at a series of future time points, and the reference pose includes at least one of a reference lateral position and a reference heading angle; solving a speed tracking control problem for tracking a planned speed to obtain the desired acceleration of the first vehicle; wherein the planned speed includes a reference speed at a series of future time points; obtaining an acceleration control quantity for the first vehicle based on at least one of the actual acceleration and the vehicle's driving resistance, and based on the desired acceleration of the first vehicle.

[0261] One embodiment of this application provides an electronic chip, including: a processor for executing computer program instructions stored in a memory, wherein when the computer program instructions are executed by the processor, the electronic chip is triggered to execute the method described in any embodiment of this application.

[0262] One embodiment of this application provides an electronic device including at least one processor and a memory coupled together. The memory is used to store computer program instructions, and the processor is used to execute the computer program instructions. When the computer program instructions are executed by the processor, the electronic device is triggered to execute the method described in any embodiment of this application.

[0263] One embodiment of this application provides a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the methods described in any embodiment of this application.

[0264] One embodiment of this application provides a computer program product, which includes a computer program that, when run on a computer, causes the computer to perform the methods described in any embodiment of this application.

[0265] Figure 7 This is a schematic diagram of a computer device provided according to one embodiment of this application. Figure 7 As shown, the computer device 20 in this embodiment includes a processor 21 and a memory 22. The memory 22 stores a computer program 23 that can run on the processor 21. When the computer program 23 is executed by the processor 21, it implements the steps in the method embodiments of this application. To avoid repetition, these steps are not described in detail here. Alternatively, when the computer program 23 is executed by the processor 21, it implements the functions of each model / unit in the device embodiments of this application. To avoid repetition, these functions are not described in detail here.

[0266] Computer device 20 includes, but is not limited to, processor 21 and memory 22. Those skilled in the art will understand that... Figure 7 This is merely an example of computer device 20 and does not constitute a limitation on computer device 20. It may include more or fewer components than shown, or combine certain components, or different components. For example, computer device may also include input / output devices, network access devices, buses, etc.

[0267] The processor 21 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or it can be any conventional processor.

[0268] The memory 22 can be an internal storage unit of the computer device 20, such as a hard disk or RAM of the computer device 20. The memory 22 can also be an external storage device of the computer device 20, such as a plug-in hard disk, Smart Media (SM) card, Secure Digital (SD) card, or FlashCard equipped on the computer device 20. Furthermore, the memory 22 can include both internal and external storage units of the computer device 20. The memory 22 is used to store the computer program 23 and other programs and data required by the computer device. The memory 22 can also be used to temporarily store data that has been output or will be output.

[0269] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, apparatuses, or units, and may be electrical, mechanical, or other forms.

[0270] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0271] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.

[0272] An integrated unit implemented as a software functional unit can be stored in a computer-readable storage medium. This software functional unit, stored in a storage medium, includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0273] In this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0274] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0275] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments of this application can be implemented using electronic hardware, computer software, or a combination of electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0276] Those skilled in the art will readily understand that, for the sake of convenience and brevity, the same or similar parts between the various embodiments of this application can be referred to mutually. For example, the specific working processes of the systems, devices, and units described in the embodiments of this application can be referred to the corresponding processes in the method embodiments of this application, and will not be repeated here.

[0277] The above description is merely a specific embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A vehicle control system, characterized in that, include: Vehicle lateral control module and vehicle longitudinal control module; The vehicle lateral control module includes a first wheel steering angle control module, and the vehicle longitudinal control module includes a speed tracking control module and an acceleration compensation module. The first wheel steering angle control module is used to iteratively solve a nonlinear unconstrained optimization problem based on the minimum principle to obtain the first wheel steering angle control quantity of the first vehicle. The first wheel is one of the front wheel and the rear wheel. The nonlinear unconstrained optimization problem is obtained by transforming the path tracking control problem for tracking the planned path using the obstacle function method and the Lagrange multiplier method. The planned path includes reference poses at a series of future time points. The reference poses include at least one of reference lateral position and reference heading angle. The speed tracking control module is used to solve the speed tracking control problem for tracking the planned speed and obtain the expected acceleration of the first vehicle; wherein, the planned speed includes reference speeds at a series of future time points; The acceleration compensation module is used to obtain the acceleration control amount of the first vehicle based on at least one of the actual acceleration of the first vehicle and the vehicle driving resistance of the first vehicle, and based on the expected acceleration of the first vehicle.

2. The vehicle control system according to claim 1, characterized in that, The vehicle lateral control module also includes a second wheel steering angle control module; The second wheel steering angle control module is used to obtain the second wheel steering angle control amount of the first vehicle based on the correlation between the front wheel steering angle and the rear wheel steering angle of the first vehicle and the first wheel steering angle control amount of the first vehicle, wherein the second wheel is the other one of the front wheel and the rear wheel; The correlation is obtained based on the maximum value of the front wheel steering angle of the first vehicle and the maximum value of the rear wheel steering angle of the first vehicle.

3. The vehicle control system according to claim 1 or 2, characterized in that, The path tracking control problem is established based on the pose error of the first vehicle relative to the Frenet coordinate system, where the origin of the Frenet coordinate system is on the planned path. The path tracking control problem is used to solve for the first wheel steering angle control quantity that satisfies the lateral control constraints and minimizes the first function. The first function includes a path tracking error optimization term and a smooth control quantity optimization term. The path tracking error optimization term is obtained based on the pose error between the vehicle trajectory and the planned path, and the smooth control quantity optimization term is obtained based on the error of the first wheel steering angle of the vehicle at adjacent time points.

4. The vehicle control system according to claim 1 or 2, characterized in that, Each solution result includes state variables reflecting pose error and control variables reflecting the steering angle of the first wheel at a series of future time points; The method of iteratively solving the nonlinear unconstrained optimization problem based on the minimum principle to obtain the first wheel steering angle control value of the first vehicle includes: If the solution result of the previous iteration exists, then based on the solution result of the previous iteration, the nonlinear unconstrained optimization problem is solved iteratively to obtain the solution result of the current iteration; If both the inner loop condition and the outer loop condition are met, then the first wheel steering angle control quantity of the first vehicle is obtained based on the control quantity obtained in the current loop; wherein, the inner loop condition includes that the error between the state quantity obtained in the previous loop and the current loop is not greater than a set error threshold, and the outer loop condition includes that the value of the first parameter in the nonlinear unconstrained optimization problem reaches a set number of times threshold. If the inner loop condition is not met, then the steps for solving the nonlinear unconstrained optimization problem are executed again. If the inner loop condition is true and the outer loop condition is false, then the value of the first parameter is increased, and the steps for solving the nonlinear unconstrained optimization problem are executed again.

5. The vehicle control system according to claim 1 or 2, characterized in that, The speed tracking control problem is derived from a longitudinal control model, which includes the correlation between the vehicle's desired acceleration, the vehicle's actual acceleration, and the parameters of the first-order inertial element. The velocity tracking control problem is used to solve for the desired acceleration that satisfies the longitudinal control constraints and minimizes the second function. The second function includes a speed tracking error, which describes the error between the reference speed and the vehicle speed predicted based on the desired acceleration.

6. The vehicle control system according to claim 1 or 2, characterized in that, The solution to the velocity tracking control problem used to track the planned velocity includes: Based on the speed tracking control problem, a third function is iteratively constructed from back to front, starting from the chronological order, for a series of future time points. The expected acceleration at each time point is obtained by minimizing the third function at each time point, and the expected acceleration of the first vehicle is the expected acceleration at the next time point. The third function at any given time point is obtained based on the speed tracking error and expected acceleration at that time point, as well as the speed tracking error at the next time point. The speed tracking error at any given time point is used to describe the error between the reference speed at that time point and the vehicle speed predicted based on the expected acceleration at the previous time point.

7. The vehicle control system according to claim 1 or 2, characterized in that, The vehicle driving resistance includes at least one of wheel steering angle resistance, slope resistance and tire rolling resistance; The wheel steering angle resistance is obtained based on the first wheel steering angle control amount of the first vehicle, the equivalent longitudinal force of the first wheel axle, and the equivalent lateral force of the first wheel axle. The equivalent longitudinal force of the first wheel axle and the equivalent lateral force of the first wheel axle are both obtained based on the first wheel steering angle control amount of the first vehicle and the speed of the first vehicle. The ramp resistance is determined based on the weight of the first vehicle and the slope of the ground where the first vehicle is located. The tire rolling resistance is obtained based on the gravity, the slope, and the set tire rolling resistance coefficient.

8. A vehicle control method, characterized in that, include: The nonlinear unconstrained optimization problem is solved iteratively based on the minimum principle to obtain the first wheel steering angle control value of the first vehicle, where the first wheel is one of the front wheel and the rear wheel; wherein, the nonlinear unconstrained optimization problem is obtained by transforming the path tracking control problem for tracking the planned path using the obstacle function method and the Lagrange multiplier method, and the planned path includes the reference pose at a series of future time points, and the reference pose includes at least one of the reference lateral position and the reference heading angle; Solve the speed tracking control problem for tracking the planned speed to obtain the desired acceleration of the first vehicle; wherein the planned speed includes reference speeds at a series of future time points; The acceleration control amount of the first vehicle is obtained based on at least one of the actual acceleration of the first vehicle and the vehicle driving resistance of the first vehicle, and based on the desired acceleration of the first vehicle.

9. An electronic device, characterized in that, The electronic device includes at least one processor coupled to a memory for storing computer program instructions and for executing the computer program instructions, wherein when the computer program instructions are executed by the processor, the electronic device is triggered to perform the method as described in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when run on a computer, causes the computer to perform the method as described in claim 8.

Images