Control Method, Device, Vehicle and Storage Medium of a Vehicle

Abstract

The present application discloses a control method, device, vehicle and storage medium for a vehicle. The method performs longitudinal control on the vehicle through at least one round of iterative process. Among them, the i-th round of iterative process includes: obtaining the actual state quantity; obtaining the terminal cost, where the terminal cost is determined based on a plurality of predicted state quantities determined in the (i-1)-th round of iterative process; the predicted state quantities include the predicted displacement and predicted speed of the vehicle; determining a planning control model based on the actual state quantity and the terminal cost; and performing longitudinal control on the vehicle based on the planning control model. Since the vehicle considers both the current actual state quantity of the vehicle and the predicted state quantities determined in the previous iterative process when determining the planning control model, the control quantity determined by the vehicle can be closer to the global optimal solution, effectively reducing the control error and overshoot of the control system and improving the longitudinal tracking accuracy of the vehicle.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle control, and more specifically, to a control method, device, vehicle, and storage medium for a vehicle. Background Art

[0002] With the popularization and development of intelligent driving technology, the longitudinal tracking control of vehicles has become the research focus of researchers. Specifically, when the upstream Motion Planning (MP) module determines the planned path, the Motion Control (MC) module in the vehicle will determine the control quantity of the vehicle based on the planned path given by the MP module to achieve the longitudinal tracking control of the vehicle.

[0003] In the existing technical solutions, the MC module will use traditional control algorithms (for example, PID control algorithm, state feedback control algorithm) to determine the control quantity of the vehicle. For example, when using the PID algorithm to determine the control quantity, the MC module will obtain the current state information of the vehicle (for example, the current vehicle speed, the current position, etc.), and input the above state information and the planned information corresponding to the planned path (for example, the planned vehicle speed, the planned position, etc.) into the PID algorithm, and then obtain the corresponding control quantity (for example, the target acceleration).

[0004] However, the traditional control algorithms only focus on the current state information of the vehicle, and problems such as tracking control lag or lead often occur in control systems with rapid state changes, which in turn leads to large errors in the longitudinal tracking control of the vehicle. Summary of the Invention

[0005] Embodiments of the present application provide a control method, device, vehicle, and storage medium for a vehicle.

[0006] In a first aspect, some embodiments of the present application provide a control method for a vehicle. The method performs longitudinal control on the vehicle through at least one round of iterative process. Among them, the i-th iterative process in at least one round of iterative process includes the following steps, where i is a positive integer: obtaining the actual state quantity, where the actual state quantity includes the actual displacement and actual speed of the vehicle; obtaining the terminal cost, where the terminal cost is determined based on a plurality of predicted state quantities determined in the (i - 1)-th iterative process and is used to characterize the driving cost required for the vehicle to travel to the target end point; the predicted state quantities include the predicted displacement and predicted speed of the vehicle; determining a planning control model based on the actual state quantity and the terminal cost; and performing longitudinal control on the vehicle based on the planning control model.

[0007] Second aspect, some embodiments of the present application further provide a control device for a vehicle. The device includes a control module, and the control module is configured to longitudinally control the vehicle through at least one round of iterative process. Specifically, the control module includes a first acquisition sub-module, a second acquisition sub-module, a determination sub-module, and a control sub-module. Among them, the first acquisition sub-module is configured to acquire actual state quantities, and the actual state quantities include the actual displacement and actual speed of the vehicle. The second acquisition sub-module is configured to acquire a terminal cost, and the terminal cost is determined based on a plurality of predicted state quantities determined in the (i-1)-th round of iterative process, and is used to characterize the driving cost required for the vehicle to travel to the target end point; the predicted state quantities include the predicted displacement and predicted speed of the vehicle. The determination sub-module is configured to determine a planning control model based on the actual state quantities and the terminal cost. The control sub-module is configured to longitudinally control the vehicle based on the planning control model.

[0008] Third aspect, some embodiments of the present application further provide a vehicle, which includes: one or more processors, a memory, and one or more applications. Among them, the one or more applications are stored in the memory and are configured to be executed by the one or more processors, and are configured to execute the above-mentioned method.

[0009] Fourth aspect, embodiments of the present application further provide a computer-readable storage medium, in which computer program instructions are stored. Among them, the computer program instructions can be called by a processor to execute the above-mentioned method.

[0010] Fifth aspect, embodiments of the present application further provide a computer program product, which when executed, implements the above-mentioned method.

[0011] The present application provides a control method, device, vehicle and storage medium for a vehicle. The control method longitudinally controls the vehicle through at least one round of iterative process. Specifically, in the i-th round of iterative process, the vehicle acquires the current actual state quantities and the terminal cost used to characterize the driving cost required for traveling to the target end point, and then longitudinally controls the vehicle based on the planning control model determined based on the terminal cost and the actual state quantities. Since the terminal cost in the present application is determined based on a plurality of predicted state quantities determined in the (i-1)-th round of iterative process, that is, when the vehicle determines the planning control model, it will simultaneously consider the current actual state quantities of the vehicle (that is, the actual displacement and actual speed) and the predicted state quantities determined in the previous iterative process, so that the control quantity determined by the vehicle can be closer to the global optimal solution, effectively reducing the control error and overshoot of the control system, and improving the longitudinal tracking accuracy of the vehicle. Description of the Drawings

[0012] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those skilled in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0013] Figure 1 The structural schematic diagram of a vehicle provided by the embodiment of the present application is shown.

[0014] Figure 2 The flowchart of a vehicle control method provided by the first embodiment of the present application is shown.

[0015] Figure 3 The flowchart of a vehicle control method provided by the second embodiment of the present application is shown.

[0016] Figure 4 The flowchart of a vehicle control method provided by the third embodiment of the present application is shown.

[0017] Figure 5 The block diagram of a vehicle control device provided by the embodiment of the present application is shown.

[0018] Figure 6 The block diagram of a vehicle provided by the embodiment of the present application is shown.

[0019] Figure 7 The block diagram of a computer-readable storage medium provided by the embodiment of the present application is shown. Detailed implementation manners

[0020] The following details the implementation manners of the present application. The examples of the implementation manners are shown in the accompanying drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end. The implementation manners described below with reference to the accompanying drawings are exemplary only for explaining the present application and should not be construed as a limitation to the present application.

[0021] To enable those skilled in the art to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present application.

[0022] The present application provides a control method, device, vehicle, and storage medium for a vehicle. The control method performs longitudinal control on the vehicle through at least one round of iterative process. Specifically, in the i-th round of iterative process, the vehicle obtains the current actual state variables and the terminal cost used to characterize the cost required to travel to the target end point, and then performs longitudinal control on the vehicle based on the planning control model determined based on the terminal cost and the actual state variables. Since the terminal cost in the present application is determined based on multiple predicted state variables determined in the (i - 1)-th round of iterative process, that is, when the vehicle determines the planning control model, it will simultaneously consider the current actual state variables of the vehicle (that is, the actual displacement and actual speed) and the predicted state variables determined in the previous iterative process, so that the control quantity determined by the vehicle can be closer to the global optimal solution, effectively reducing the control error and overshoot of the control system and improving the longitudinal tracking accuracy of the vehicle.

[0023] To facilitate a detailed description of the solution of the present application, the application environment in the embodiments of the present application will be introduced below with reference to the accompanying drawings. Please refer to Figure 1 , the control method for the vehicle provided in the embodiments of the present application is applied to the vehicle 100. The vehicle 100 refers to a means of transportation driven or towed by a power device for people to ride or for transporting goods, including but not limited to sedans, sports utility vehicles (SUVs), multi-purpose vehicles (MPVs), and the like. Specifically, the vehicle 100 may include a console 110 and an execution system 120.

[0024] The central console 110 is the control center of the vehicle 100, which is used to process the data information obtained by the vehicle 100 during driving and generate control instructions for controlling the vehicle 100. In this embodiment, the central console 110 may include a planning control module. When the central console 110 determines the current actual state variables (i.e., actual displacement and actual speed) and the terminal cost, based on this planning control module, the corresponding control quantity can be determined and sent to the execution system 120. The execution system 120 works based on this control quantity to achieve the tracking control of the planned path, that is, to achieve the longitudinal control of the vehicle 100. In some possible embodiments, the planning control module may also be set in a server communicatively connected to the vehicle 100. When the central console 110 needs to determine the control quantity, it can send the current actual state variables and the terminal cost of the vehicle 100 to the server in real time and receive the control quantity determined by the server based on the planning control module. Among them, the server can be a single server, a server cluster composed of multiple servers, or a cloud computing service center. In some possible embodiments, the server may be the background server corresponding to the intelligent driving function in the central console 110. Specifically, the planning control module and the terminal cost are introduced in detail in the following method embodiments.

[0025] In this embodiment, a path planning module is also pre-stored in the central console 110. The path planning module can calculate the planned path based on the obtained target starting point and target ending point. Among them, the planned path can be composed of multiple planned path points, and each planned path point can include information such as planned displacement, planned speed, and planned acceleration. Specifically, a path planning algorithm can be set in the path planning module, and the path planning algorithm can include search algorithms (such as Dijkstra algorithm, A* algorithm, Weighted A* algorithm, etc.), and this embodiment does not make specific limitations on this. Similarly, the path planning module can also be set in a server communicatively connected to the vehicle 100. When the central console 110 needs to determine the planned path, it sends the target starting point and target ending point to the server and receives the planned path determined by the server based on the path planning module.

[0026] The execution system 120 is electrically connected to the central console 110 and can work based on the control quantity output by the central console 110 to control the driving state of the vehicle 100. In this embodiment, the execution system 120 can receive the target acceleration determined by the central console 110 based on the planning control model and adjust the rotational speed of the wheels based on this target acceleration to achieve the acceleration or deceleration of the vehicle 100, that is, to achieve the longitudinal control of the vehicle 100.

[0027] Please refer to Figure 2 , Figure 2Schematically shown is a control method for a vehicle provided by the first embodiment of the present application. The control method performs longitudinal control on the vehicle through at least one round of iterative process. Wherein, the i-th iterative process in at least one round of iterative process includes steps S210 to S240, and i is a positive integer.

[0028] Step S210, obtain the actual state quantity.

[0029] In this embodiment, the actual state quantity includes the actual displacement and actual speed of the vehicle. As an implementation manner, when the central console turns on the automatic driving function and has determined the planned path, it obtains the current position and actual speed of the vehicle every preset time period, and calculates the distance between the current position and the target starting point based on a preset distance calculation formula, and then determines this distance as the actual displacement of the vehicle.

[0030] Among them, the preset distance calculation formula can be the Euclidean distance calculation formula, the Manhattan distance calculation formula, etc. The preset time period can be the default value in the central console, or can be adjusted by R & D personnel based on the longitudinal control accuracy of the vehicle. Specifically, the higher the longitudinal control accuracy, the shorter the preset time period, so that the number of times of obtaining the actual state quantity is more, that is, the number of iterations of longitudinal control is more.

[0031] As an implementation manner, the central console can obtain the current position of the vehicle and the target starting point based on the Global Positioning System (GPS), and obtain the actual speed of the vehicle through an in-vehicle speed sensor (for example, a magnetoelectric speed sensor, a Hall speed sensor). This embodiment does not make specific limitations on this.

[0032] Step S220, obtain the terminal cost.

[0033] In this embodiment, the terminal cost is determined based on multiple predicted state quantities determined in the (i - 1)-th iterative process, and is used to characterize the driving cost required for the vehicle to travel to the target end point. The predicted state quantities include the predicted displacement and predicted speed of the vehicle. Since in this round of iterative process, the central console has not preset multiple predicted state quantities, the central console will determine the terminal cost based on multiple predicted state quantities determined in the previous round (that is, the (i - 1)-th round) of iterative process.

[0034] It should be noted here that the multiple predicted state variables determined in the (i - 1)-th iteration refer to the multiple predicted displacements and predicted speeds predicted when the vehicle travels to the target end point based on the historical state variables (i.e., the actual state variables obtained in the (i - 1)-th iteration). At this time, the central console combines the multiple planned displacements and planned speeds on the planned path, calculates the first difference between the multiple predicted displacements and the multiple planned displacements in the (i - 1)-th iteration and the second difference between the multiple predicted speeds and the multiple planned speeds in the (i - 1)-th iteration, and determines the sum of the first difference and the second difference as the terminal cost. Specifically, the calculation method of the terminal cost is introduced in the following embodiments.

[0035] Step S230, determine a planning control model based on the actual state variables and the terminal cost.

[0036] In this embodiment, the planning control model can be an optimization model determined based on the actual state variables and the terminal cost. Therefore, this planning control model not only involves the current actual state variables of the vehicle (i.e., the actual displacement and actual speed), but also involves the predicted state variables determined in the previous iteration process, so that when the central console optimizes and solves this planning control model subsequently, the determined control quantity can be closer to the global optimal solution, thereby effectively reducing the control error and overshoot of the control system and improving the longitudinal tracking accuracy of the vehicle. Specifically, the determination process of the planning control model is introduced in detail in the following embodiments.

[0037] Step S240, perform longitudinal control on the vehicle based on the planning control model.

[0038] In this embodiment, the central console can determine the control quantity corresponding to the longitudinal control (for example, the target acceleration) based on the planning control model, and send this control quantity to the execution system of the vehicle. The execution system works based on this control quantity to achieve longitudinal control of the vehicle. For example, the execution system can receive the target acceleration determined by the central console based on the planning control model, and adjust the rotation speed of the wheels based on this target acceleration to achieve acceleration or deceleration of the vehicle, that is, to achieve longitudinal control of the vehicle.

[0039] This embodiment provides a control method for a vehicle. Since the terminal cost in this method is determined based on the multiple predicted state variables determined in the (i - 1)-th iteration, that is, when the vehicle determines the planning control model, it will simultaneously consider the current actual state variables of the vehicle (i.e., the actual displacement and actual speed) and the predicted state variables determined in the previous iteration process, so that the control quantity determined by the vehicle can be closer to the global optimal solution, effectively reducing the control error and overshoot of the control system and improving the longitudinal tracking accuracy of the vehicle.

[0040] Please refer to Figure 3 ,Figure 3 Schematically shown is a control method for a vehicle provided by the second embodiment of the present application. This control method longitudinally controls the vehicle through at least one round of iterative process. In this embodiment, the determination process of the planning control model and the determination process of the control quantity corresponding to the longitudinal control are introduced in detail. Specifically, the i-th iterative process in at least one round of iterative process includes steps S310 to S390, where i is a positive integer.

[0041] Step S310: Obtain the actual state quantity.

[0042] Specifically, the specific implementation manner of step S310 can refer to the detailed description in step S210 and will not be elaborated here one by one.

[0043] Step S320: Obtain the terminal cost.

[0044] In some embodiments, when i is equal to 1, that is, when the central console is in the first round of iterative process, step S320 may include steps S3210 to S3250.

[0045] Step S3210: Obtain M planned state quantities and M planned accelerations determined by a preset path planning algorithm.

[0046] In this embodiment, when the central console receives an instruction to turn on the autonomous driving function, it will obtain the target starting point and the target ending point of the vehicle. Among them, the target starting point may be the current position of the vehicle, and the target ending point may be the target position input by the user received by the central console. When the central console determines the target starting point and the target ending point, it will determine a planned path from the target starting point to the target ending point based on a preset path planning algorithm. Specifically, the planned path includes multiple path planning points, and each path planning point may include a planned state quantity and a planned acceleration. Among them, the planned state quantity includes planned displacement and planned speed. When the central console determines multiple path planning points, it can store the planned state quantity and the planned acceleration corresponding to each path planning point in the local memory. Subsequently, when the central console needs the planned state quantity and the planned acceleration, it can directly read the data in the local memory.

[0047] Step S3230: Based on a preset closed-loop control algorithm, obtain M predicted state quantities and M predicted accelerations.

[0048] In this embodiment, since it is the first-round iterative control of the center console, the center console does not obtain any predicted state variables and predicted accelerations. In this case, the center console will obtain M predicted state variables and M predicted accelerations based on the closed-loop control algorithm, that is, multiple predicted values corresponding to the vehicle driving to the target end point. Among them, the preset closed-loop control algorithm includes proportional-integral-derivative control algorithm (PID algorithm) or state feedback control algorithm (LQR algorithm), etc.

[0049] Step S3250, determine the terminal cost based on the M planned state variables, M planned accelerations, M predicted state variables, and M predicted accelerations.

[0050] In this embodiment, the center console can respectively determine the first difference between the M planned state variables and the M predicted state variables and determine the second difference between the M planned accelerations and the M predicted accelerations, and determine the sum of the first difference and the second difference as the terminal cost. Specifically, when the center console determines the M planned state variables, M planned accelerations, M predicted state variables, and M predicted accelerations, the terminal cost is calculated by the following formula.

[0051]

[0052] Among them, t represents the current time. In this embodiment, t takes the value of 0. ∞ represents the end time. x t represents the predicted state variable corresponding to time t, x t_ref represents the planned state variable corresponding to time t, u t represents the predicted state variable corresponding to time t, u t_ref represents the planned acceleration corresponding to time t. Q and R are weight matrices respectively, and the values of Q and R are both greater than 0. That is, the terminal cost is positively correlated with the difference between the planned state variable and the predicted state variable of the vehicle, and the terminal cost is positively correlated with the difference between the planned acceleration and the predicted acceleration of the vehicle. Among them, Q and R can be default parameters or can be adjusted by R & D personnel. Specifically, when the difference between the planned state variable and the predicted state variable of the vehicle has a greater impact on the terminal cost, the value of Q is increased so that Q is greater than R. When the difference between the planned acceleration and the predicted acceleration of the vehicle has a greater impact on the terminal cost, the value of R is increased so that R is greater than Q. The factor that has a greater impact on the terminal cost (the difference between the planned state variable and the predicted state variable of the vehicle, or the difference between the planned acceleration and the predicted acceleration of the vehicle) can be set according to the actual working conditions or can be set artificially. It should be noted here that in the case where i is greater than 1, k + N in the above formula represents the time corresponding to the (N + 1)-th state point after the state point corresponding to time k. Specifically, when solving the terminal cost corresponding to the i-th round, x in the above formula tis the predicted state quantity determined in the (i-1)-th round.

[0053] This embodiment gives the calculation method of the terminal cost in the first-round iteration process. Since it is the first-round iteration, in the case of lacking predicted state quantity and predicted acceleration data, the central console will determine the above-mentioned planned quantity based on the preset closed-loop control algorithm, thereby ensuring the smooth progress of the iteration process.

[0054] In some embodiments, when i is greater than 1, the central console can determine the terminal cost through the following formula.

[0055]

[0056]

[0057] where x represents the state quantity corresponding to different times t, j represents the current iteration number, represents the terminal cost required to travel from the x state to the target end point in the j-th round of iteration process (that is, j = i-1), represents the weight, represents the terminal cost determined in the previous iteration process, and the specific calculation formula can refer to step S3250.

[0058] Step S330, obtain N planned state quantities and the corresponding N planned accelerations determined by the preset path planning algorithm.

[0059] Specifically, the specific implementation method of obtaining the planned state quantity and the planned acceleration can refer to the detailed introduction in step S3210. It should be noted here that N represents the number of path planning points closest to the current position of the vehicle at the current position. Specifically, N can be the default value in the central console, or can be adjusted by the R & D personnel based on the longitudinal control accuracy of the vehicle or the available computing resources of the central console. Among them, the higher the longitudinal control accuracy of the vehicle, the larger the value of N; the more available computing resources of the central console, the larger the value of N.

[0060] Step S340, obtain the control sequence.

[0061] In this embodiment, the control sequence includes N predicted accelerations. Among them, the N predicted accelerations represent the accelerations predicted by the central console after the vehicle travels at different times at the current position. Specifically, the N predicted accelerations are arranged in sequence according to their corresponding times. As an implementation manner, the central console can determine the default value as the initial value of the control sequence. The central console can also use the control sequence determined in the historical iteration process as the control sequence in this round of iteration process. This embodiment does not specifically limit the determination method of the control sequence.

[0062] Step S350: Determine N - 1 predicted state variables based on the control sequence and the actual state variables.

[0063] In this embodiment, the center console can determine N - 1 predicted state variables based on a preset vehicle dynamics model. The vehicle dynamics model can be pre - stored in the center console. Specifically, the vehicle dynamics model is as follows.

[0064] x t+1 = Ax t + Bu t .

[0065] Where x t represents the predicted state variable corresponding to time t. When t is the current time, x t is the actual state variable obtained by the center console. Specifically, where s represents the predicted displacement corresponding to time t, and v represents the predicted speed corresponding to time t. When t is the current time, s is the actual displacement and v is the actual speed. u t represents the predicted acceleration corresponding to time t, that is, the input variable. x t+1 represents the predicted state variable corresponding to time t + 1. A and B are state matrix and input matrix with known parameters respectively. Specifically,

[0066] Here, the determination process of the predicted state variables is introduced with N taking the value of 3. At this time, the control sequence includes 3 predicted accelerations, that is, u1, u2, and u3. Since x1 is a known quantity, that is, the actual state variable, the 2 predicted state variables (x2 and x3) that the center console needs to determine can be expressed by the following formulas.

[0067] x2 = Ax1 + Bu1;

[0068] x3 = Ax2 + Bu2 = A(Ax1 + Bu1)+Bu2.

[0069] Therefore, when x1, A, and B are all known, the 2 predicted state variables can be expressed by the control sequence.

[0070] Step S360: Determine the planned control model based on the actual state variables, N - 1 predicted state variables, N planned state variables, N planned accelerations, the control sequence, and the terminal cost.

[0071] In this embodiment, the planning control model is represented by a cost function. The center console can respectively determine the third difference between the actual state quantity, N - 1 predicted state quantities and N planned state quantities, and determine the fourth difference between N planned accelerations and the control sequence, and determine the sum of the third difference, the fourth difference and the terminal cost as the cost function. Specifically, the cost function corresponding to the planning control model is as follows.

[0072]

[0073]

[0074] Among them, J represents the target cost corresponding to the cost function, k represents the current moment, L is the value interval, x t represents the predicted state quantity corresponding to the moment t. Among them, when t = k, x k is the actual state quantity. x t_ref represents the planned state quantity corresponding to the moment t, u t represents the predicted state quantity corresponding to the moment t, u t_ref represents the planned acceleration corresponding to the moment t. Q and R are respectively weight matrices, and the values of Q and R are both greater than 0. That is, the target cost is positively correlated with the difference between the planned state quantity and the predicted state quantity of the vehicle, and the target cost is positively correlated with the difference between the planned acceleration and the predicted acceleration of the vehicle. Among them, Q and R can be default parameters or can be adjusted by R & D personnel. Specifically, when the difference between the planned state quantity and the predicted state quantity of the vehicle has a greater impact on the target cost, the value of Q is increased so that Q is greater than R. When the difference between the planned acceleration and the predicted acceleration of the vehicle has a greater impact on the target cost, the value of R is increased so that R is greater than Q. The factor that has a greater impact on the target cost (the difference between the planned state quantity and the predicted state quantity of the vehicle, or the difference between the planned acceleration and the predicted acceleration of the vehicle) can be set according to the actual working conditions or can be set artificially. Q final is the terminal cost. X and U are respectively the constraint spaces of the state quantity and the input quantity.

[0075] Therefore, in the case where the predicted state quantity can be represented based on the control sequence, and the planned state quantity, the planned acceleration and the terminal cost are all known, the cost function in this embodiment is an optimization function for solving the control sequence.

[0076] Step S370, optimize and solve the control sequence in the planning control model to determine the target control sequence.

[0077] In this embodiment, the center console can optimize and solve the control sequence based on a preset iterative optimization algorithm. When the number of iterations is greater than a specified number or the target cost is less than a specified cost, the optimization and solution of the control sequence are completed, and the solved control sequence is determined as the target control sequence. Specifically, the iterative optimization algorithm can be the Newton method, the gradient descent algorithm, and so on. The number of iterations and the specified cost can be default values in the center console, or can be adjusted by R & D personnel based on the optimization accuracy of the control sequence. Specifically, the higher the optimization accuracy of the control sequence, the greater the number of iterations and the smaller the specified cost. This embodiment does not make specific limitations on this.

[0078] Step S380: Determine the first value in the target control sequence as the target acceleration.

[0079] In this embodiment, the target acceleration represents the expected value of the acceleration. Since the N predicted accelerations included in the control sequence in this embodiment are arranged in sequence according to their corresponding moments. Therefore, the first value in the target control sequence is the predicted acceleration corresponding to the driving moment closest to the current moment, and the center console determines this predicted acceleration as the target acceleration.

[0080] Step S390: Perform longitudinal control on the vehicle based on the target acceleration.

[0081] In this embodiment, the center console can send the target acceleration to the execution system of the vehicle, and the execution system works based on this target acceleration to achieve longitudinal control of the vehicle.

[0082] This embodiment provides a control method for a vehicle. This method details the determination process of the planning control model and the determination process of the control quantity corresponding to the longitudinal control. Since the terminal cost in this method is determined based on multiple predicted state quantities determined in the (i - 1)-th iteration process, that is, when the vehicle determines the planning control model, it will simultaneously consider the current actual state quantities of the vehicle (that is, the actual displacement and actual speed) and the predicted state quantities determined in the previous iteration process, so that the control quantity determined by the vehicle can be closer to the global optimal solution, effectively reducing the control error and overshoot of the control system and improving the longitudinal tracking accuracy of the vehicle.

[0083] Please refer to Figure 4 , Figure 4 which schematically shows a control method for a vehicle provided in the second embodiment of the present application. This control method performs longitudinal control on the vehicle through at least one round of iterative process. Among them, the i-th iterative process in at least one round of iterative process includes steps S410 to S480, and i is a positive integer.

[0084] Step S410: Obtain the actual state quantity.

[0085] Specifically, the specific implementation manner of step S410 may refer to the detailed description in step S210, which will not be elaborated here one by one.

[0086] Step S420, obtain the set of historical state variables.

[0087] In this embodiment, the set of historical state variables includes multiple predicted state variables of the vehicle traveling from the target starting point to the target ending point during the previous i - 1 rounds of iteration. Specifically, step S420 may include steps S4210 to S4230.

[0088] Step S4210, obtain multiple predicted state variables during the previous i - 1 rounds of iteration.

[0089] In this embodiment, the central console can determine multiple predicted state variables during the previous i - 1 rounds of iteration based on the iteration results during the historical iteration process. Specifically, the multiple predicted state variables can be represented by the following formula.

[0090]

[0091] Where i is the number of iterations, j represents the number of iterations that have been performed, is the predicted state variable corresponding to time t during the i - th round of iteration. Among them, when t = 0, is the predicted state variable corresponding to the target starting point during the i - th round of iteration. When t = ∞, is the predicted state variable corresponding to the target ending point during the i - th round of iteration. U represents the union operation. Therefore, SS in the above formula j represents the set corresponding to multiple predicted state variables pre - determined for the vehicle traveling from the target starting point to the target terminal during the previous i - 1 rounds of iteration.

[0092] In some embodiments, the central console can initially set the set SS j to an empty set. After each round of iteration is completed, add the multiple predicted state variables determined in the current round into the set SS j . Therefore, when the central console executes step S4210, it can directly read the data in the set SS j to determine multiple predicted state variables during the previous i - 1 rounds of iteration.

[0093] However, in the case where i is equal to 1, that is, during the first round of iteration, the set SS j is an empty set. In this case, step S4210 includes step S4212.

[0094] Step S4212, based on a preset closed - loop control algorithm, obtain multiple predicted state variables.

[0095] In this embodiment, the center console obtains multiple predicted state variables based on a preset closed-loop control algorithm. The preset closed-loop control algorithm includes a proportional-integral-derivative control algorithm (PID algorithm) or a state feedback control algorithm (LQR algorithm). After determining the multiple predicted state variables, the center console adds the multiple predicted state variables to the above set SS j , to ensure the smooth execution of subsequent steps.

[0096] Step S4230: Construct a convex set corresponding to the multiple predicted state variables, and determine the convex set as the historical state variable set.

[0097] In this embodiment, the convex set can be represented by the following formula.

[0098]

[0099] where conv(SS j ) represents the convex set corresponding to the set SS j , cn is the cardinality of the set SS j , x i is a state variable in the set SS j , and a i is the weight corresponding to the state variable x i . Since the multiple state variables in the set SS j are discrete values, by linearizing the multiple discrete values and uniformly representing them as a convex set, a feasible value range corresponding to the state variables (including the actual state variables and the predicted state variables) can be determined.

[0100] Step S430: Obtain the terminal cost when the value corresponding to the actual state variable belongs to the historical state variable set.

[0101] In this embodiment, the terminal cost can be represented by the following formula.

[0102]

[0103] where, when the value corresponding to the actual state variable belongs to the historical state variable set, that is, x ∈ CS j , it indicates that there is no large deviation between the obtained actual state variable and the previously obtained predicted state variables. At this time, it means that the vehicle is running smoothly. In this case, the center console obtains the terminal cost The terminal cost is characterized by the terminal cost function . The specific determination process of the terminal cost function can refer to the detailed description in step 320 and will not be elaborated here.

[0104] Conversely, when the value corresponding to the actual state quantity does not belong to the set of historical state quantities, it indicates that a sensor in the vehicle has failed or the driving state of the vehicle has fluctuated significantly. In this case, if the terminal cost function determined from the historical terminal cost is still obtained, it will cause the vehicle to be unable to achieve good longitudinal control in this round of iteration. In this case, the center console does not execute the subsequent steps S440 to S480, but determines the target acceleration through a preset closed-loop control algorithm (for example, PID algorithm, LQR algorithm), thereby achieving longitudinal control of the vehicle.

[0105] Step S440: Determine the planning control model based on the actual state quantity and the terminal cost.

[0106] Step S450: Perform longitudinal control on the vehicle based on the planning control model.

[0107] For the specific implementation manners of step S440 and step S450, reference can be made to the detailed descriptions in steps S330 to S390, which will not be elaborated here one by one.

[0108] Step S460: Obtain the first terminal cost value.

[0109] In this embodiment, the first terminal cost value represents the value of the terminal cost function corresponding to the zero moment in the i-th round of iteration. Specifically, the center console obtains the terminal cost function determined in this round (i.e., the i-th round), and assigns the parameter t in this terminal cost function to 0 to determine the first terminal cost value.

[0110] Step S470: Obtain the second terminal cost value.

[0111] In this embodiment, the second terminal cost value represents the value of the terminal cost function corresponding to the zero moment in the (i - 1)-th round of iteration. Specifically, the center console obtains the terminal cost function determined in the previous round (i.e., the (i - 1)-th round), and assigns the parameter t in this terminal cost function to 0 to determine the second terminal cost value.

[0112] Step S480: End the iteration process when the difference between the second terminal cost value and the first terminal cost value is less than or equal to the specified difference.

[0113] In this embodiment, the center console ends the iteration process when the difference between the second terminal cost value and the first terminal cost value is less than or equal to the specified difference. Specifically, the calculation formula for the difference between the second terminal cost value and the first terminal cost value can be as follows.

[0114]

[0115] Where is the second terminal cost value, is the first terminal cost value, and ε is a specified difference. Specifically, the specified difference can be a default value in the center console, or can be adjusted by the R & D personnel based on the control accuracy of longitudinal control. That is, the higher the control accuracy of longitudinal control, the smaller the specified difference. In some possible embodiments, the specified difference can be equal to 0.

[0116] In some embodiments, when the difference between the second terminal cost value and the first terminal cost value in the center console is greater than the specified difference, the next round of iterative process is started, that is, it starts to execute again from step S410.

[0117] This embodiment provides a control method for a vehicle. Since the terminal cost in this method is determined based on multiple predicted state quantities determined in the (i - 1)-th round of iterative process, that is, when the vehicle determines the planning control model, it will simultaneously consider the current actual state quantities of the vehicle (that is, the actual displacement and actual speed) and the predicted state quantities determined in the previous iterative process, so that the control quantity determined by the vehicle can be closer to the global optimal solution, effectively reducing the control error and overshoot of the control system and improving the longitudinal tracking accuracy of the vehicle.

[0118] Please refer to Figure 5 , Figure 5 schematically shows a control device 500 for a vehicle provided by an embodiment of the present application. The control device 500 for the vehicle includes a control module 510, and the control module 510 is used to perform longitudinal control on the vehicle through at least one round of iterative process. Specifically, the control module 510 includes a first acquisition sub-module 5100, a second acquisition sub-module 5110, a determination sub-module 5120, and a control sub-module 5130. Among them, the first acquisition sub-module 5100 is used to acquire actual state quantities, and the actual state quantities include the actual displacement and actual speed of the vehicle. The second acquisition sub-module 5110 is used to acquire terminal cost, and the terminal cost is determined based on multiple predicted state quantities determined in the (i - 1)-th round of iterative process, and is used to characterize the driving cost required for the vehicle to travel to the target end point; the predicted state quantities include the predicted displacement and predicted speed of the vehicle. The determination sub-module 5120 is used to determine a planning control model based on the actual state quantities and the terminal cost. The control sub-module 5130 is used to perform longitudinal control on the vehicle based on the planning control model.

[0119] In some embodiments, the determination sub-module 5120 is further used to acquire N planning state quantities and the corresponding N planned accelerations determined by a preset path planning algorithm; acquire a control sequence, and the control sequence includes N predicted accelerations; determine N - 1 predicted state quantities based on the control sequence and the actual state quantities; determine a planning control model based on the actual state quantities, N - 1 predicted state quantities, N planning state quantities, N planned accelerations, the control sequence, and the terminal cost.

[0120] In some embodiments, the N predicted accelerations included in the control sequence are arranged in sequence according to their corresponding times. The control sub-module 5130 is further configured to optimize and solve the control sequence in the planning control model to determine the target control sequence; determine the first value in the target control sequence as the target acceleration, where the target acceleration represents the expected value of the acceleration; and perform longitudinal control on the vehicle based on the target acceleration.

[0121] In some embodiments, the second acquisition sub-module 5110 is further configured to acquire a set of historical state quantities, where the set of historical state quantities includes a plurality of predicted state quantities of the vehicle traveling from the target starting point to the target ending point during the previous i-1 rounds of iteration; and acquire the terminal cost in the case where the value corresponding to the actual state quantity belongs to the set of historical state quantities.

[0122] In some embodiments, the second acquisition sub-module 5110 is further configured to acquire a plurality of predicted state quantities during the previous i-1 rounds of iteration; construct a convex set corresponding to the plurality of predicted state quantities, and determine the convex set as the set of historical state quantities.

[0123] In some embodiments, when i is equal to 1, the second acquisition sub-module 5110 is further configured to acquire a plurality of predicted state quantities based on a preset closed-loop control algorithm, where the preset closed-loop control algorithm includes a proportional integral derivative control algorithm or a state feedback control algorithm.

[0124] In some embodiments, when i is equal to 1, the second acquisition sub-module 5110 is further configured to acquire M planned state quantities and M planned accelerations determined by a preset path planning algorithm; the M planned state quantities and the M planned accelerations represent a plurality of planned values corresponding to the vehicle traveling to the target ending point; acquire M predicted state quantities and M predicted accelerations based on a preset closed-loop control algorithm, where the preset closed-loop control algorithm includes a proportional integral derivative control algorithm or a state feedback control algorithm; and determine the terminal cost based on the M planned state quantities, the M planned accelerations, the M predicted state quantities, and the M predicted accelerations.

[0125] In some embodiments, the terminal cost is characterized by a terminal cost function. The control module 510 further includes a third acquisition sub-module (not shown in the figure), a fourth acquisition sub-module (not shown in the figure), and an end module (not shown in the figure). Among them, the third acquisition sub-module is configured to acquire a first terminal cost value, where the first terminal cost value represents the value of the terminal cost function corresponding to the zero moment during the i-th round of iteration. The fourth acquisition sub-module is configured to acquire a second terminal cost value, where the second terminal cost value represents the value of the terminal cost function corresponding to the zero moment during the (i-1)-th round of iteration. The end module is configured to end the iteration process when the difference between the second terminal cost value and the first terminal cost value is less than or equal to a specified difference.

[0126] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0127] In several embodiments provided by the present application, the coupling between modules can be electrical, mechanical or other forms of coupling.

[0128] In addition, in each embodiment of the present application, the various functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0129] This embodiment provides a control device for a vehicle. Since the terminal cost in this device is determined based on multiple predicted state quantities determined in the (i - 1)-th iteration process, that is, when determining the planning control model of the vehicle, the actual state quantities of the vehicle at present (that is, the actual displacement and actual speed) and the predicted state quantities determined in the previous iteration process are considered simultaneously, so that the control quantity determined by the vehicle can be closer to the global optimal solution, effectively reducing the control error and overshoot of the control system and improving the longitudinal tracking accuracy of the vehicle.

[0130] Please refer to Figure 6 , Figure 6 which schematically shows that the embodiments of the present application further provide a vehicle 600, and the vehicle 600 includes: one or more processors 610, a memory 620, and one or more application programs. Among them, the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to execute the methods described in the above embodiments.

[0131] The processor 610 may include one or more processing cores. The processor 610 is connected to various parts within the entire battery management system through various interfaces and circuits. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 620, and by invoking data stored in the memory 620, it performs various functions of the battery management system and processes data. Optionally, the processor 610 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 610 may integrate a combination of one or several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing display content; the modem is used to process wireless communication. It can be understood that the above modem may not be integrated into the processor 610 and may be implemented separately through a communication chip.

[0132] The memory 620 may include a random access memory (RAM) and may also include a read-only memory (ROM). The memory 620 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 620 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for implementing at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above various method embodiments, etc. The data storage area may also store data created during the use of the electronic device diagram (such as phone book, audio and video data, chat record data), etc.

[0133] Please refer to Figure 7 , Figure 7 which schematically shows that the embodiment of the present application further provides a computer-readable storage medium 700. Computer program instructions 710 are stored in the computer-readable storage medium 700, and the computer program instructions 710 can be called by the processor to execute the methods described in the above embodiments.

[0134] The computer-readable storage medium 700 can be, for example, a flash memory, an electrically erasable programmable read-only memory (EEPROM), an electrically programmable read-only memory (EPROM), a hard disk, or a read-only memory (ROM). Optionally, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium 700 has a storage space for computer program instructions 710 that execute any of the method steps in the above-described method. These computer program instructions 710 can be read from or written to one or more computer program products.

[0135] The above are only the preferred embodiments of the present application, and do not impose any formal restrictions on the present application. Although the present application has been disclosed above with preferred embodiments, it is not intended to limit the present application. Any person skilled in the art can make some modifications or decorations equivalent to equivalent embodiments by using the technical content disclosed above within the scope of the technical solution of the present application. However, any brief modifications, equivalent changes, and decorations made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application still fall within the scope of the technical solution of the present application.

Claims

1. A control method for a vehicle, characterized in that, The method includes: Performing longitudinal control on the vehicle through at least one round of iterative process; wherein, the i-th round of iterative process in at least one round of the iterative process includes the following steps, where i is a positive integer: Obtaining the actual state quantity, where the actual state quantity includes the actual displacement and actual speed of the vehicle; Obtaining the terminal cost, where the terminal cost is determined based on a plurality of predicted state quantities determined in the (i - 1)-th round of iterative process and is used to characterize the driving cost required for the vehicle to travel to the target end point; the predicted state quantities include the predicted displacement and predicted speed of the vehicle; Determining a planning control model based on the actual state quantity and the terminal cost; wherein, determining the planning control model based on the actual state quantity and the terminal cost includes: obtaining N planning state quantities and corresponding N planned accelerations determined by a preset path planning algorithm; obtaining a control sequence, where the control sequence includes N predicted accelerations; the N predicted accelerations included in the control sequence are arranged in sequence according to their corresponding moments; determining N - 1 predicted state quantities based on the control sequence and the actual state quantity; determining the planning control model based on the actual state quantity, N - 1 predicted state quantities, N planning state quantities, N planned accelerations, the control sequence, and the terminal cost; Performing longitudinal control on the vehicle based on the planning control model; wherein, performing longitudinal control on the vehicle based on the planning control model includes: optimizing and solving the control sequence in the planning control model to determine a target control sequence; determining the first value in the target control sequence as the target acceleration, where the target acceleration represents the expected value of the acceleration; performing longitudinal control on the vehicle based on the target acceleration.

2. The method according to claim 1, wherein The obtaining the terminal cost includes: Obtaining a set of historical state quantities, where the set of historical state quantities includes a plurality of predicted state quantities of the vehicle traveling from the target starting point to the target end point in the previous i - 1 rounds of iterative process; Obtaining the terminal cost when the value corresponding to the actual state quantity belongs to the set of historical state quantities.

3. The method according to claim 2, characterized in that, The obtaining the set of historical state quantities includes: Obtaining a plurality of the predicted state quantities in the previous i - 1 rounds of iterative process; Constructing a convex set corresponding to the plurality of predicted state quantities and determining the convex set as the set of historical state quantities.

4. The method according to claim 3, characterized in that When i is equal to 1, the obtaining a plurality of the predicted state quantities in the previous i - 1 rounds of iterative process includes: Obtaining a plurality of the predicted state quantities based on a preset closed-loop control algorithm, where the preset closed-loop control algorithm includes a proportional-integral-derivative control algorithm or a state feedback control algorithm.

5. The method according to claim 1, characterized in that, When i is equal to 1, the obtaining the terminal cost includes: Obtaining M planning state quantities and M planned accelerations determined by a preset path planning algorithm; the M planning state quantities and M planned accelerations represent a plurality of planned values corresponding to the vehicle traveling to the target end point. Based on a preset closed-loop control algorithm, obtain M predicted state variables and M predicted accelerations, where the preset closed-loop control algorithm includes a proportional-integral-derivative control algorithm or a state feedback control algorithm; Based on the M planned state variables, M planned accelerations, M predicted state variables, and M predicted accelerations, determine the terminal cost.

6. The method according to claim 1, characterized in that, The terminal cost is characterized by a terminal cost function. After longitudinally controlling the vehicle based on the planned control model, the method further includes: Obtain a first terminal cost value, where the first terminal cost value represents the value of the terminal cost function at zero time in the i-th iteration process; Obtain a second terminal cost value, where the second terminal cost value represents the value of the terminal cost function at zero time in the (i - 1)-th iteration process; End the iteration process when the difference between the second terminal cost value and the first terminal cost value is less than or equal to a specified difference.

7. A control device for a vehicle, characterized in that, The device includes: A control module for longitudinally controlling the vehicle through at least one iteration process; The control module includes a first acquisition sub-module, a second acquisition sub-module, a determination sub-module, and a control sub-module; The first acquisition sub-module is used to acquire actual state variables, where the actual state variables include the actual displacement and actual speed of the vehicle; The second acquisition sub-module is used to acquire a terminal cost, where the terminal cost is determined based on a plurality of predicted state variables determined in the (i - 1)-th iteration process and is used to characterize the driving cost required for the vehicle to travel to the target end point; the predicted state variables include the predicted displacement and predicted speed of the vehicle; The determination sub-module is used to determine a planned control model based on the actual state variables and the terminal cost; specifically, the determination sub-module is used to obtain N planned state variables and corresponding N planned accelerations determined by a preset path planning algorithm; obtain a control sequence, where the control sequence includes N predicted accelerations; the N predicted accelerations included in the control sequence are arranged in sequence according to their corresponding times; based on the control sequence and the actual state variables, determine N - 1 predicted state variables; based on the actual state variables, N - 1 predicted state variables, N planned state variables, N planned accelerations, the control sequence, and the terminal cost, determine the planned control model; The control sub-module is used to longitudinally control the vehicle based on the planned control model; specifically, the control sub-module is used to optimize and solve the control sequence in the planned control model to determine a target control sequence; determine the first value in the target control sequence as the target acceleration, where the target acceleration represents the expected value of the acceleration; longitudinally control the vehicle based on the target acceleration.

8. A vehicle, characterized in that, Includes: One or more processors; A memory; One or more applications, where one or more of the applications are stored in the memory and are configured to be executed by one or more of the processors and are configured to execute the method according to any one of claims 1 - 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, and the computer program instructions can be called by a processor to execute the method according to any one of claims 1-6.

Images