Model prediction-based lane keeping method and apparatus, vehicle and storage medium

Abstract

The present application relates to the technical field of vehicles, in particular to a model prediction-based lane keeping method and apparatus, a vehicle and a storage medium. The method comprises: on the basis of an acquired front wheel steering angle of a vehicle, lateral distance deviation between a vehicle center position and a road center position, and yaw angle deviation of the vehicle, constructing a target lateral dynamics model; on the basis of the target lateral dynamics model, using a preset exponentially decaying expectation function to optimize the lateral distance deviation and the yaw angle deviation, so as to obtain a trajectory planning path of the vehicle, and simultaneously using a preset prediction model to track the trajectory planning path, so as to obtain a lane keeping request torque of the vehicle; and acquiring a steering wheel torque of the vehicle, and, on the basis of the steering wheel torque and the lane keeping request torque, obtaining a target torque of the vehicle, so as to control the vehicle to perform lane keeping. Thus, the present application solves the problems that due to whole-vehicle modeling, the lane deviation correction accuracy is low and better control effects can only be achieved in some scenarios, etc.

Description

Lane keeping method and device based on model prediction, vehicle and storage medium

[0001] Cross-reference to related applications

[0002] The present application is based on the Chinese patent application No. 202410741330.X, filed on June 7, 2024, and claims priority to the Chinese patent application, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD

[0003] The present application relates to the technical field of vehicles, in particular to a lane keeping method and device based on model prediction, a vehicle and a storage medium. BACKGROUND

[0004] With the progress of automobile electrification, intelligentization and networking, more and more vehicles are equipped with ADAS (Advanced Driving Assistance System) automatic driving functions to help users improve driving safety. Among them, the lane keeping strategy is an important part of the lateral auxiliary strategy, which is a safety guarantee for driving. When the vehicle deviates from the center line of the lane, the system triggers the function to start, and the vehicle will return to the center line of the lane with a certain trajectory, and the heading angle will be consistent with the lane, thus completing the lane keeping to ensure driving safety and avoid driving risks.

[0005] In related technologies, the state quantity acquisition and model establishment of the lane keeping strategy are realized by artificial intelligence algorithms. Since some required state quantities are not easy to obtain and have low stability, and the lack of data sources will lead to only estimation during lane keeping, which will reduce the lane correction accuracy. At the same time, the related algorithm modeling is biased towards the whole vehicle rather than lateral modeling, and can only obtain good control effect in part of the scene, thus there is a driving safety risk, which needs to be solved urgently.

[0006] SUMMARY

[0007] The present application provides a lane keeping method and device based on model prediction, a vehicle and a storage medium to solve the problems of low lane correction accuracy caused by whole vehicle modeling and only good control effect in part of the scene.

[0008] The first aspect embodiment of the present application provides a lane keeping method based on model prediction, comprising the following steps:

[0009] obtaining a front wheel steering angle of the vehicle, a lateral distance deviation between a center position of the vehicle and a center position of a road where the vehicle is located, and a yaw angle deviation of the vehicle, and constructing a target lateral dynamics model according to the front wheel steering angle, the lateral distance deviation, and the yaw angle deviation;

[0010] optimizing the lateral distance deviation and the yaw angle deviation based on the target lateral dynamics model by using a preset exponential decay expectation function, to obtain a trajectory planning path of the vehicle, and tracking the trajectory planning path by using a preset prediction model, to obtain a lane keeping request torque of the vehicle;

[0011] obtaining a steering wheel torque of the vehicle, obtaining a target torque of the vehicle according to the steering wheel torque and the lane keeping request torque, and controlling the vehicle to keep the lane according to the target torque.

[0012] According to an embodiment of the present application, before obtaining a front wheel steering angle of the vehicle, a lateral distance deviation between a center position of the vehicle and a center position of a road where the vehicle is located, and a yaw angle deviation of the vehicle, the method further comprises:

[0013] obtaining driving information of the vehicle, steering wheel input information, and road information where the vehicle is located;

[0014] performing data cleaning on the driving information, the steering wheel input information, and the road information where the vehicle is located based on a preset data cleaning strategy, to obtain cleaned driving information, steering wheel input information, and road information.

[0015] According to an embodiment of the present application, the optimizing the lateral distance deviation and the yaw angle deviation based on the target lateral dynamics model by using a preset exponential decay expectation function, to obtain a trajectory planning path of the vehicle, comprises:

[0016] obtaining an expectation function of the lateral distance deviation and an expectation function of the yaw angle deviation;

[0017] optimizing the lateral distance deviation and the yaw angle deviation based on the preset exponential decay expectation function by using the expectation function of the lateral distance deviation and the expectation function of the yaw angle deviation respectively, to obtain the trajectory planning path of the vehicle.

[0018] According to an embodiment of the present application, the tracking the trajectory planning path by using a preset prediction model comprises:

[0019] Based on the preset predictive control equations, target prediction functions are constructed for the front wheel steering angle, the lateral distance deviation, and the yaw angle deviation. Then, based on the target prediction functions, the first constraint condition corresponding to the front wheel steering angle, the second constraint condition corresponding to the lateral distance deviation, and the third constraint condition corresponding to the yaw angle deviation are obtained.

[0020] Based on the first constraint, the second constraint, and the third constraint, the optimal path control sequence of the vehicle is determined, a preset prediction model of the vehicle is obtained, and the trajectory planning path is tracked based on the preset prediction model.

[0021] According to one embodiment of this application, obtaining the steering wheel torque of the vehicle includes:

[0022] The driver's hand force, the vehicle's speed, and the steering wheel angle are obtained.

[0023] The driver's hand force torque, the vehicle's driving speed damping torque, and the steering wheel angle return torque are generated based on a preset torque mapping relationship.

[0024] The steering wheel torque of the vehicle is obtained by summing the hand force torque, the damping torque, and the return torque.

[0025] According to one embodiment of this application, the preset exponential decay expectation function is: e y (k+i│k)=(1-e -λi )e y (k);

[0026] Among them, e y The lateral distance deviation between the vehicle's center position and the road center position, (1-e) -λi ) represents the expected trajectory coefficient, k represents the current period, i represents the i-th prediction period, (k+i|k) represents the state predicted in the k+i-th period during the k-th period, and e -λi For the coefficients of the exponential decay function, e y (k) is the first value of the lateral distance deviation in the kth cycle.

[0027] According to the lane keeping method based on model prediction provided in the embodiments of the present application, a target lateral dynamics model is constructed according to the obtained front wheel steering angle of the vehicle, lateral distance deviation of the vehicle from the center position of the road, and yaw angle deviation of the vehicle; based on the target lateral dynamics model, the lateral distance deviation and the yaw angle deviation are optimized by using a preset exponential decay expectation function, so as to obtain a trajectory planning path of the vehicle, and meanwhile, the trajectory planning path is tracked by using a preset prediction model, so as to obtain a lane keeping request torque of the vehicle; the steering wheel torque of the vehicle is obtained, and the target torque of the vehicle is obtained in combination with the lane keeping request torque, and then the vehicle is controlled to keep the lane. In this way, the problem of low lane correction accuracy caused by whole vehicle modeling and the problem of good control effect in only part of scenarios are solved.

[0028] The second aspect of the present application provides a lane keeping device based on model prediction, comprising:

[0029] An acquisition module is configured to acquire a front wheel steering angle of a vehicle, a lateral distance deviation of a center position of the vehicle from a center position of a road where the vehicle is located, and a yaw angle deviation of the vehicle, and construct a target lateral dynamics model according to the front wheel steering angle, the lateral distance deviation, and the yaw angle deviation;

[0030] A path tracking module is configured to optimize the lateral distance deviation and the yaw angle deviation by using a preset exponential decay expectation function based on the target lateral dynamics model, so as to obtain a trajectory planning path of the vehicle, and meanwhile, track the trajectory planning path by using a preset prediction model, so as to obtain a lane keeping request torque of the vehicle;

[0031] A control module is configured to acquire a steering wheel torque of the vehicle, obtain a target torque of the vehicle according to the steering wheel torque and the lane keeping request torque, and control the vehicle to keep the lane according to the target torque.

[0032] According to an embodiment of the present application, before acquiring a front wheel steering angle of a vehicle, a lateral distance deviation of a center position of the vehicle from a center position of a road where the vehicle is located, and a yaw angle deviation of the vehicle, the acquisition module further comprises:

[0033] A first acquisition unit is configured to acquire driving information of the vehicle, steering wheel input information, and road information where the vehicle is located;

[0034] A data cleaning unit is configured to perform data cleaning on the driving information, the steering wheel input information, and the road information where the vehicle is located based on a preset data cleaning strategy, so as to obtain cleaned driving information, steering wheel input information, and road information.

[0035] According to one embodiment of the present application, the path tracking module comprises:

[0036] A second acquisition unit is configured to acquire the expected function of the lateral distance deviation and the expected function of the yaw angle deviation.

[0037] An optimization unit is configured to optimize the lateral distance deviation and the yaw angle deviation based on the preset exponential decay expected function, the expected function of the lateral distance deviation and the expected function of the yaw angle deviation respectively, to obtain the trajectory planning path of the vehicle.

[0038] According to one embodiment of the present application, the path tracking module comprises:

[0039] A third acquisition unit is configured to construct the target prediction function of the front wheel steering angle, the lateral distance deviation and the yaw angle deviation based on a preset prediction control equation, and obtain the first constraint condition corresponding to the front wheel steering angle, the second constraint condition corresponding to the lateral distance deviation and the third constraint condition corresponding to the yaw angle deviation according to the target prediction function.

[0040] A determination unit is configured to determine the optimal path control sequence of the vehicle based on the first constraint condition, the second constraint condition and the third constraint condition, obtain the preset prediction model of the vehicle, and track the trajectory planning path based on the preset prediction model.

[0041] According to one embodiment of the present application, the control module comprises:

[0042] A fourth acquisition unit is configured to acquire the driver's hand force, the driving speed of the vehicle and the steering wheel steering angle.

[0043] A generation unit is configured to generate the hand force torque of the driver's hand force, the damping torque of the driving speed of the vehicle and the return torque of the steering wheel steering angle based on a preset torque mapping relationship.

[0044] A torque adding unit is configured to add the hand force torque, the damping torque and the return torque to obtain the steering wheel torque of the vehicle.

[0045] According to one embodiment of the present application, the preset exponential decay expected function is: e y (k+i│k)=(1-e -λi )e y (k);

[0046] Wherein, e y is the lateral distance deviation of the center position of the vehicle and the road center position where the vehicle is located, (1-e -λi) is an expected trajectory coefficient, k is a current period, i is i prediction periods, (k+i│k) is a state of a (k+i)th period predicted in a kth period, e -λi is an exponential decay function coefficient, e y (k) is a first value of a lateral distance deviation of a kth period.

[0047] According to the lane keeping device based on model prediction provided in the embodiments of the present application, a target lateral dynamics model is constructed according to the obtained front wheel steering angle of the vehicle, the lateral distance deviation of the vehicle from the road center position, and the yaw angle deviation of the vehicle; based on the target lateral dynamics model, the lateral distance deviation and the yaw angle deviation are optimized by using a preset exponential decay expected function, so as to obtain a trajectory planning path of the vehicle, and meanwhile, the trajectory planning path is tracked by using a preset prediction model, so as to obtain a lane keeping request torque of the vehicle; the steering wheel torque of the vehicle is obtained, and the lane keeping request torque is combined to obtain a target torque of the vehicle, and then the vehicle is controlled to keep the lane. In this way, the problem that the lane correction accuracy is low and good control effect can only be obtained in part of scenes due to the whole vehicle modeling is solved.

[0048] The third aspect of the embodiments of the present application provides a vehicle, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the lane keeping method based on model prediction as described in the above embodiments.

[0049] The fourth aspect of the embodiments of the present application provides a computer readable storage medium, which stores computer instructions for causing the computer to execute the lane keeping method based on model prediction as described in the above embodiments.

[0050] The additional aspects and advantages of the present application will be partially given in the following description, partially become obvious from the following description, or be known by the practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0051] The above and / or additional aspects and advantages of the present application will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

[0052] FIG. 1 is a flowchart of a lane keeping method based on model prediction according to an embodiment of the present application;

[0053] FIG. 2 is a system block diagram of lane keeping according to an embodiment of the present application;

[0054] FIG. 3 is a schematic diagram of an exponential decay function according to an embodiment of the present application;

[0055] FIG. 4 is an execution schematic diagram of an EPS (Electric Power Steering) module according to an embodiment of the present application;

[0056] FIG. 5 is a block diagram of a lane keeping device based on model prediction according to an embodiment of the present application;

[0057] FIG. 6 is a structural schematic diagram of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0058] Embodiments of the present application are described in detail below with reference to the accompanying drawings, in which the same or similar components have the same or similar designations and functions throughout. The embodiments described below are exemplary, and are intended to explain the present application, and are not to be understood as limiting the present application.

[0059] A lane keeping method based on model prediction, a device, a vehicle and a storage medium are described below with reference to the accompanying drawings. In view of the low lane correction accuracy caused by the whole vehicle modeling and the good control effect only in some scenarios mentioned in the background art, the present application provides a lane keeping method based on model prediction. In the method, a target lateral dynamics model is constructed according to the front wheel steering angle of the vehicle, the lateral distance deviation between the vehicle and the road center position, and the yaw angle deviation of the vehicle. Based on the target lateral dynamics model, the lateral distance deviation and the yaw angle deviation are optimized by using a preset exponential decay expectation function, and a trajectory planning path of the vehicle is obtained. At the same time, the trajectory planning path is tracked by using a preset prediction model, and a lane keeping request torque of the vehicle is obtained. The steering wheel torque of the vehicle is obtained, and the target torque of the vehicle is obtained by combining the lane keeping request torque, and the vehicle is controlled to keep the lane. Thus, the problems of low lane correction accuracy caused by whole vehicle modeling and good control effect only in some scenarios are solved.

[0060] Specifically, FIG. 1 is a flowchart of a lane keeping method based on model prediction provided by an embodiment of the present application.

[0061] As shown in FIG. 1, the lane keeping method based on model prediction includes the following steps:

[0062] In step S101, the front wheel steering angle of the vehicle, the lateral distance deviation between the center position of the vehicle and the road center position where the vehicle is located, and the yaw angle deviation of the vehicle are obtained, and a target lateral dynamics model is constructed according to the front wheel steering angle, the lateral distance deviation and the yaw angle deviation.

[0063] According to one embodiment of the present application, before acquiring the front wheel steering angle of the vehicle, the lateral distance deviation of the center position of the vehicle and the road center position where the vehicle is located, and the yaw angle deviation of the vehicle, the method further comprises: acquiring driving information of the vehicle, steering wheel input information, and road information where the vehicle is located; and performing data cleaning on the driving information, the steering wheel input information, and the road information where the vehicle is located based on a preset data cleaning strategy to obtain cleaned driving information, cleaned steering wheel input information, and cleaned road information.

[0064] The preset data cleaning strategy can be a data cleaning strategy selected by a person skilled in the art according to actual test requirements, or a relatively optimal data cleaning strategy obtained through a limited number of tests, and is not specifically limited here.

[0065] Specifically, to avoid the vehicle deviating from the lane during the user driving process and thus causing traffic accidents such as vehicle scratching and collision, the embodiments of the present application can effectively utilize the kinematic information and dynamic information of the vehicle, combine the road information, utilize the model prediction method, design the trajectory planning and trajectory tracking equation, and finally realize accurate calculation of the EPS total torque, so as to make the vehicle return to the right and return to the center of the lane to ensure the driving safety of the user.

[0066] Specifically, as shown in FIG. 2, the embodiments of the present application first need to acquire driving information of the vehicle, steering wheel input information, and road information where the vehicle is located, wherein the driving information of the vehicle can be the driving state of the vehicle, such as the lateral speed, the longitudinal speed of the vehicle, and the lateral position of the vehicle from the lane line; the steering wheel input information can be the steering wheel steering angle and the torque; and the road information where the vehicle is located can be the road curvature and the curvature radius; secondly, since the above-mentioned information mainly comes from the body controller, the camera, the radar, the GPS (Global Positioning System), and the like, the sources are wide, and thus it is necessary to perform data cleaning on the acquired driving information, steering wheel input information, and road information where the vehicle is located based on a preset data cleaning strategy, for example, to eliminate abnormal data information (extreme values deviating by 2%, 98% quantile), to fill in empty numerical values according to the interpolation method, to perform data homodyning (so that the data has uniformity and stability), and the like, so as to obtain cleaned driving information of the vehicle, cleaned steering wheel input information, and cleaned road information, and to take the cleaned driving information, cleaned steering wheel input information, and cleaned road information as an important precondition for subsequent data processing.

[0067] Further, the embodiment of the present application mainly studies the response characteristics of the vehicle when the vehicle is transversely corrected, therefore, the lateral performance of the vehicle needs to be tested, the embodiment of the present application fully considers the stress characteristics, lateral motion law and tire lateral stiffness of the vehicle, and has good response characteristics under high-speed and low-speed working conditions, therefore, the body stiffness, air resistance can be ignored, the motion in the z-axis direction is not considered, the front wheel angle of the vehicle is obtained as the input of the vehicle system, the lateral distance deviation of the center position of the vehicle and the road center position and the yaw angle deviation of the vehicle are obtained as the state variables of the vehicle system, and the kinematic equations are established along the longitudinal direction (set as x direction), the lateral direction (set as y direction) and the rotation around the Z axis to keep the lane, as shown in the following formula:

[0068] wherein, δ f is the front wheel angle of the vehicle, φ is the body yaw angle, C lf is the front wheel longitudinal stiffness, C cf is the front wheel lateral stiffness, C Ir is the rear wheel longitudinal stiffness, C cr is the rear wheel lateral stiffness, s f is the front wheel slip ratio, s r is the rear wheel slip ratio, I z is the moment of inertia, a is the front axle distance, b is the rear axle distance, is the longitudinal acceleration, is the lateral velocity, is the yaw angular velocity, is the lateral acceleration, is the yaw angular acceleration, and m is the vehicle mass.

[0069] Further, in order to simplify the research problem and reduce the calculation complexity, it is assumed that the front wheel angle is consistent when the front wheel of the vehicle rotates, when the vehicle intends to correct, the expected yaw angular velocity is determined by the road radius R and the longitudinal velocity , that is

[0070] Let e y be the lateral distance deviation (i.e. lateral error) of the center position of the vehicle and the road center position, e φ be the yaw angle deviation (i.e. the angle error between the vehicle heading angle and the road tangent), e y and e φ , are state variables, the front wheel angle δ fFor the input variable, the target lateral dynamics model is established based on the selected state variable and the input variable, thereby further improving the lane correction accuracy of the vehicle, as shown in the following formula:

[0071] wherein, is the lateral error change rate, is the heading angle error change rate, m is the vehicle mass, and the state variable

[0072] In step S102, the lateral distance deviation and the yaw angle deviation are optimized using a preset exponential decay expectation function based on the target lateral dynamics model, to obtain a trajectory planning path of the vehicle, and the trajectory planning path is tracked using a preset prediction model to obtain a lane keeping request torque of the vehicle.

[0073] According to an embodiment of the present application, the trajectory planning path of the vehicle is obtained by optimizing the lateral distance deviation and the yaw angle deviation using a preset exponential decay expectation function based on the target lateral dynamics model, comprising: obtaining an expectation function of the lateral distance deviation and an expectation function of the yaw angle deviation; based on the preset exponential decay expectation function, the lateral distance deviation and the yaw angle deviation are optimized using the expectation function of the lateral distance deviation and the expectation function of the yaw angle deviation, respectively, to obtain the trajectory planning path of the vehicle.

[0074] According to an embodiment of the present application, the trajectory planning path is tracked using a preset prediction model, comprising: constructing a target prediction function of the front wheel steering angle, the lateral distance deviation and the yaw angle deviation based on a preset prediction control equation, and obtaining a first constraint condition corresponding to the front wheel steering angle, a second constraint condition corresponding to the lateral distance deviation and a third constraint condition corresponding to the yaw angle deviation according to the target prediction function; based on the first constraint condition, the second constraint condition and the third constraint condition, the optimal path control sequence of the vehicle is determined, to obtain the preset prediction model of the vehicle, and the trajectory planning path is tracked based on the preset prediction model.

[0075] wherein, the preset exponential decay expectation function, the preset prediction control equation and the preset prediction model can be selected by those skilled in the art according to actual test requirements, which are not specifically limited here.

[0076] Specifically, in the lane keeping system, it is necessary to consider that the vehicle body is directed to be consistent with the current road tangent direction as much as possible, and the vehicle does not press the lane line, i.e. the vehicle mass center should be located in the middle of the road as much as possible, thereby the optimization function can be determined to obtain the error of the lateral distance deviation and the yaw angle deviation two state variables, and the error is calculated in the form of square sum to obtain the total error, whose formula is shown in the following formula: s.t.δ f,min ≤δf ≤ δ f,max ;

[0077] wherein k1, k2 are polynomial coefficients, is the sum of square of lateral distance deviation, is the sum of square of yaw angle deviation, δ f,min is the minimum front wheel steering angle, δ f,max is the maximum front wheel steering angle, s.t. is the constraint condition.

[0078] Further, as shown in FIG. 3, the embodiment of the present application needs to plan a trajectory for the driving condition of the vehicle, so as to control the vehicle to return to the center line of the road with the optimal path, such as constant speed offset, circular arc based trajectory, forward and reverse trapezoidal based trajectory, etc. In order to make the calculation process more effective and stable, and accelerate the convergence speed of the model, therefore, the embodiment of the present application adopts a preset exponential decay expectation function to obtain the trajectory planning path of the vehicle, so as to make the state variable tend to the optimal value, wherein the preset exponential decay expectation function, i.e. the expected reference equation of the lateral distance deviation e y between the center position of the vehicle and the road center position where the vehicle is located, is as follows: e y (k+i|k) = (1-e -λi )e y (k);

[0079] wherein e y is the lateral distance deviation between the center position of the vehicle and the road center position where the vehicle is located, (1-e -λi ) is the expected trajectory coefficient, k is the current period, i is the i th time interval in the current period, (k+i|k) is the state after the i th time interval is predicted in the k th period, e -λi is the exponential decay function coefficient, e y (k) is the first value of the lateral distance deviation in the k th period.

[0080] It should be noted that the present application can be written in the form of a standard vector X ref (k+i|k) = φ i X(k), wherein φ i is the expected trajectory coefficient diagonal matrix, X(k) is the first value of all state variables X in the k th period, X ref (k+i|k) is the expected value of all state variables X after the i th time interval in the k th period.

[0081] Specifically, the embodiment of the present application first obtains an expected function of the lateral distance deviation and an expected function of the yaw angle deviation based on the lateral distance deviation and the yaw angle deviation of the vehicle respectively; secondly, the expected function of the lateral distance deviation and the expected function of the yaw angle deviation are used to optimize the lateral distance deviation and the yaw angle deviation based on the preset exponential decay expected function, so that the lateral distance deviation and the yaw angle deviation tend to be optimal values, and the effectiveness and stability thereof are improved, and then the trajectory planning path of the vehicle is further obtained; since the purpose of the lane keeping of the present application is to make the state variables converge as soon as possible, and the front wheel steering angle δ f and the increment Δδ f of the front wheel steering angle are as small as possible, therefore, finally, the embodiment of the present application needs to construct a target prediction function and list the corresponding constraint conditions, solve the optimal path control sequence, and take the first value in the solved optimal path control sequence as the input value, and discard the other values in the optimal path control sequence, that is, the embodiment of the present application needs to construct the target prediction function of the front wheel steering angle, the lateral distance deviation and the yaw angle deviation based on the preset prediction control equation, and obtain the first constraint condition corresponding to the front wheel steering angle, the second constraint condition corresponding to the lateral distance deviation and the third constraint condition corresponding to the yaw angle deviation according to the target prediction function, wherein the first constraint condition, the second constraint condition and the third constraint condition can be directly calculated by the following formula: s.t.δ f,min ≤δ f ≤δ f,max ; Δδ f,min ≤Δδ f ≤Δδ f,max ;

[0082] wherein X(k+i│k) and X ref (k+i│k) are the state at (k+1) time based on the prediction at k time, the ideal state; Δδ f (k+i) is the input increment at (k+1) time; Q is the weight coefficient of the system state error, and R is the weight coefficient of the control input increment.

[0083] Further, the embodiment of the present application can determine the optimal path control sequence of the vehicle based on the first constraint condition, the second constraint condition and the third constraint condition, and in the next cycle, the process is recycled. The embodiment of the present application first performs a predictive conversion on the performance execution, constructs a cost function and a constraint condition for a prediction time domain, so as to convert the cost function into a predictive type, thereby obtaining a preset prediction model of the vehicle, for example, an MPC (model predictive control) prediction model, and tracking the trajectory planning path based on the MPC prediction model, and further obtaining the lane keeping request torque for controlling the EPS.

[0084] It should be noted that the first constraint condition, the second constraint condition and the third constraint condition calculated by the embodiments of the present application are all wheel rotation angles δ f In the physical range.

[0085] In step S103, the steering wheel torque of the vehicle is obtained, the target torque of the vehicle is obtained according to the steering wheel torque and the lane keeping request torque, and the vehicle is controlled to keep the lane according to the target torque.

[0086] According to an embodiment of the present application, the steering wheel torque of the vehicle is obtained, including: obtaining the driver's hand force, the driving speed of the vehicle and the steering wheel rotation angle; generating the hand force torque of the driver's hand force, the damping torque of the driving speed of the vehicle and the return torque of the steering wheel rotation angle based on the preset torque mapping relationship; and adding the hand force torque, the damping torque and the return torque to obtain the steering wheel torque of the vehicle.

[0087] The preset torque mapping relationship can be selected by a person skilled in the art according to actual test requirements, and is not limited here.

[0088] Specifically, as shown in FIG. 4, the system architecture of the lane keeping function is shown, where Tq+ is a positive torque and Tq- is a negative torque. After obtaining the optimal path control sequence with the constraint condition based on the above embodiments of the present application, the lane keeping request torque of the vehicle is finally obtained, and the actual running data of the vehicle is also obtained, for example, the driver's hand force, the driving speed of the vehicle and the steering wheel rotation angle. The hand force torque of the driver's hand force, the damping torque of the driving speed of the vehicle and the return torque of the steering wheel rotation angle are generated based on the preset torque mapping relationship in the table through the basic boost curve corresponding to the driver's hand force, the damping force curve corresponding to the driving speed of the vehicle and the return torque curve corresponding to the steering wheel rotation angle through the table lookup method. The hand force torque, the damping torque and the return torque are added to obtain the steering wheel torque of the vehicle. At this time, the lane keeping request torque of the EPS is converted into the lane keeping request torque of the EPS through the gear angle controller by the angle request value output by the data processing module through the lane keeping request torque obtained above, and the steering wheel torque and the lane keeping request torque are calculated by the EPS. The target torque of the vehicle, that is, the total output torque of the motor, is obtained, so as to control the vehicle to keep the lane according to the target torque.

[0089] In summary, the above embodiments of the present application can have the following beneficial effects:

[0090] (1) The embodiments of the present application use an exponential decay function as the lane keeping path, thereby improving the convergence speed, effectiveness and stability of the model;

[0091] (2) The embodiment of the application adopts a model predictive control method, selects a front wheel steering angle as an input, a lateral distance deviation, a heading angle deviation, a lateral error change rate, and a heading angle error change rate as state variables, converts the problem into a quadratic form, solves an optimal problem with constraints, and thus improves trajectory correction accuracy.

[0092] According to the lane keeping method based on model prediction, the front wheel steering angle of the vehicle, the lateral distance deviation of the vehicle from the road center position, and the yaw angle deviation of the vehicle are obtained, a target lateral dynamics model is constructed, the lateral distance deviation and the yaw angle deviation are optimized based on the target lateral dynamics model by using a preset exponential decay expectation function, a trajectory planning path of the vehicle is obtained, the trajectory planning path is tracked by using a preset prediction model, a lane keeping request torque of the vehicle is obtained, a steering wheel torque of the vehicle is obtained, a target torque of the vehicle is obtained in combination with the lane keeping request torque, and the vehicle is controlled to keep the lane. Thus, the problem of low lane correction accuracy caused by whole vehicle modeling and the problem of only obtaining good control effect in some scenarios are solved.

[0093] Secondly, the lane keeping device based on model prediction is described with reference to the accompanying drawings.

[0094] FIG. 5 is a block schematic diagram of the lane keeping device based on model prediction according to the embodiment of the application.

[0095] As shown in FIG. 5, the lane keeping device based on model prediction 10 includes an acquisition module 100, a path tracking module 200, and a control module 300.

[0096] The acquisition module 100 is configured to acquire a front wheel steering angle of a vehicle, a lateral distance deviation of a center position of the vehicle from a road center position where the vehicle is located, and a yaw angle deviation of the vehicle, and construct a target lateral dynamics model according to the front wheel steering angle, the lateral distance deviation, and the yaw angle deviation.

[0097] The path tracking module 200 is configured to optimize the lateral distance deviation and the yaw angle deviation based on the target lateral dynamics model by using a preset exponential decay expectation function, obtain a trajectory planning path of the vehicle, track the trajectory planning path by using a preset prediction model, and obtain a lane keeping request torque of the vehicle.

[0098] The control module 300 is configured to acquire a steering wheel torque of the vehicle, obtain a target torque of the vehicle according to the steering wheel torque and the lane keeping request torque, and control the vehicle to keep the lane according to the target torque.

[0099] According to one embodiment of the present application, before acquiring the front wheel steering angle of the vehicle, the lateral distance deviation of the center position of the vehicle and the road center position where the vehicle is located, and the yaw angle deviation of the vehicle, the acquisition module 100 further comprises:

[0100] The first acquisition unit is configured to acquire driving information of the vehicle, steering wheel input information, and road information where the vehicle is located.

[0101] The data cleaning unit is configured to perform data cleaning on the driving information, the steering wheel input information, and the road information where the vehicle is located based on a preset data cleaning strategy, to obtain cleaned driving information, cleaned steering wheel input information, and cleaned road information.

[0102] According to one embodiment of the present application, the path tracking module 200 comprises:

[0103] The second acquisition unit is configured to acquire an expected function of the lateral distance deviation and an expected function of the yaw angle deviation.

[0104] The optimization unit is configured to optimize the lateral distance deviation and the yaw angle deviation based on the expected functions of the lateral distance deviation and the yaw angle deviation, respectively, based on a preset exponential decay expected function, to obtain a trajectory planning path of the vehicle.

[0105] According to one embodiment of the present application, the path tracking module 200 comprises:

[0106] The third acquisition unit is configured to construct target prediction functions of the front wheel steering angle, the lateral distance deviation, and the yaw angle deviation based on a preset prediction control equation, and to obtain a first constraint condition corresponding to the front wheel steering angle, a second constraint condition corresponding to the lateral distance deviation, and a third constraint condition corresponding to the yaw angle deviation according to the target prediction functions.

[0107] The determination unit is configured to determine an optimal path control sequence of the vehicle based on the first constraint condition, the second constraint condition, and the third constraint condition, to obtain a preset prediction model of the vehicle, and to track the trajectory planning path based on the preset prediction model.

[0108] According to one embodiment of the present application, the control module 300 comprises:

[0109] The fourth acquisition unit is configured to acquire the driver's hand force, the driving speed of the vehicle, and the steering wheel rotation angle.

[0110] The generation unit is configured to generate a hand torque of the driver's hand force, a damping torque of the driving speed of the vehicle, and a return torque of the steering wheel rotation angle based on a preset torque mapping relationship.

[0111] The torque adding unit is configured to add the hand torque, the damping torque, and the return torque to obtain a steering wheel torque of the vehicle.

[0112] According to one embodiment of the present application, the preset exponential decay expectation function is: e y (k+i│k)=(1-e -λi )e y (k);

[0113] wherein e y is a lateral distance deviation of a center position of the vehicle from a road center position, (1-e -λi ) is an expectation trajectory coefficient, k is a current period, i is i prediction periods, (k+i│k) is a state of the k+i period predicted in the k period, e -λi is an exponential decay function coefficient, and e y (k) is a first value of the lateral distance deviation in the k period.

[0114] According to the lane keeping device based on model prediction provided by the embodiments of the present application, a target lateral dynamics model is constructed according to the acquired front wheel steering angle of the vehicle, the lateral distance deviation of the vehicle from the road center position, and the yaw angle deviation of the vehicle; based on the target lateral dynamics model, the lateral distance deviation and the yaw angle deviation are optimized by using a preset exponential decay expectation function, so as to obtain a trajectory planning path of the vehicle, and the trajectory planning path is tracked by using a preset prediction model, so as to obtain a lane keeping request torque of the vehicle; the steering wheel torque of the vehicle is acquired, the lane keeping request torque is combined to obtain a target torque of the vehicle, and the vehicle is controlled to keep the lane. Thus, the problem of low lane correction accuracy caused by whole vehicle modeling and the problem of only obtaining good control effect in some scenarios are solved.

[0115] FIG. 6 is a structural schematic diagram of a vehicle provided by an embodiment of the present application. The vehicle can include:

[0116] a memory 601, a processor 602, and a computer program stored in the memory 601 and executable on the processor 602.

[0117] The processor 602 implements the lane keeping method based on model prediction provided in the above embodiments when executing the program.

[0118] Further, the vehicle further includes:

[0119] a communication interface 603 for communication between the memory 601 and the processor 602.

[0120] The memory 601 is used to store the computer program executable on the processor 602.

[0121] The memory 601 can include a high-speed RAM memory, and can also include a non-volatile memory, for example, at least one disk memory.

[0122] If the memory 601, the processor 602 and the communication interface 603 are implemented independently, the communication interface 603, the memory 601 and the processor 602 can be connected to each other through a bus and complete communication between each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, only one thick line is used in FIG. 6, but it does not mean that there is only one bus or only one type of bus.

[0123] Optionally, in a specific implementation, if the memory 601, the processor 602 and the communication interface 603 are integrated on a chip, the memory 601, the processor 602 and the communication interface 603 can complete communication between each other through an internal interface.

[0124] The processor 602 can be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0125] The embodiment also provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the model prediction based lane keeping method as above.

[0126] In the description of the application, reference to "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" means that a particular feature, structure, material, or characteristic being described is included in at least one embodiment or example of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples. Moreover, the terminology used in the description is for the purpose of describing the particular versions only and is not intended to be limiting. The use of particular terms to describe the particular versions should not be used to limit the scope of the application to the particular innovative features presented because the novel concepts and embodiments presented herein are capable of other implementations.

[0127] Furthermore, the terms "first", "second", and the like, do not denote any quantity or order but are used as labels for naming various elements. Hence, these terms are used interchangeably and do not denote or imply any relative importance or any particular order of the elements so named. Thus, the features labeled with "first" and "second" can explicitly or implicitly include at least one of the features. In the description of the application, the term "N" means at least two, such as two, three, etc., unless otherwise specifically stated.

[0128] Any process or method descriptions or blocks in flow charts or otherwise described herein represent embodiments which can be managed as one or more modules, segments, or portions of code which include one or more executable instructions for implementing specific logic functions or steps, and the preferred embodiments of this application also include the possibility that these one or more modules, segments, or portions of code can be implemented by firmware which manages the described functions in the manner as described above, or by hardware, or by any combination of hardware and software. The description of the application should be understood to include all possible combinations of hardware and software described herein, including a combination that does not include software but only hardware, a combination that includes software but not hardware, and a combination that includes both hardware and software.

[0129] The logic and / or steps represented in the flowcharts and / or described herein, for example, can be considered as a sequence of instructions to implement logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, processor- based system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be a computer- readable storage medium or a computer-readable signal medium. The computer- readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include the following: an electrical connection having one or more wires (electrical connections), a portable computer diskette (a magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example, via optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and stored in a computer memory.

[0130] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. As such, if implemented in hardware, and in another embodiment, any of the following technologies, known in the art, or their combinations can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), and the like.

[0131] Those skilled in the art can understand that all or part of the steps carried out by the above-mentioned embodiment methods can be completed by programs instructing related hardware, and the programs can be stored in a computer-readable storage medium. When the programs are executed, they include one or a combination of the steps of the method embodiments.

[0132] In addition, each of the functional units in the various embodiments of the present application can be integrated in one processing module, or each of the units can be physically present separately, or two or more units can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software functional module. When the integrated module is realized in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0133] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A model prediction based lane keeping method, characterized in that, The method comprises the following steps: obtaining a front wheel steering angle of a vehicle, a lateral distance deviation between a center position of the vehicle and a center position of a road where the vehicle is located, and a yaw angle deviation of the vehicle, and constructing a target lateral dynamics model according to the front wheel steering angle, the lateral distance deviation and the yaw angle deviation; based on the target lateral dynamics model, optimizing the lateral distance deviation and the yaw angle deviation by using a preset exponential decay expectation function to obtain a trajectory planning path of the vehicle, and tracking the trajectory planning path by using a preset prediction model to obtain a lane keeping request torque of the vehicle; obtaining a steering wheel torque of the vehicle, obtaining a target torque of the vehicle according to the steering wheel torque and the lane keeping request torque, and controlling the vehicle to keep the lane according to the target torque.

2. The method of claim 1, wherein, Before obtaining the front wheel steering angle of the vehicle, the lateral distance deviation between the center position of the vehicle and the center position of the road where the vehicle is located, and the yaw angle deviation of the vehicle, the method further comprises: obtaining driving information of the vehicle, steering wheel input information and road information where the vehicle is located; based on a preset data cleaning strategy, performing data cleaning on the driving information, the steering wheel input information and the road information where the vehicle is located to obtain cleaned driving information, steering wheel input information and road information.

3. The method of claim 1, wherein, The method of optimizing the lateral distance deviation and the yaw angle deviation by using the preset exponential decay expectation function based on the target lateral dynamics model to obtain the trajectory planning path of the vehicle comprises: obtaining an expectation function of the lateral distance deviation and an expectation function of the yaw angle deviation; based on the preset exponential decay expectation function, optimizing the lateral distance deviation and the yaw angle deviation by using the expectation function of the lateral distance deviation and the expectation function of the yaw angle deviation respectively to obtain the trajectory planning path of the vehicle.

4. The method of claim 1, wherein, The method of tracking the trajectory planning path by using the preset prediction model comprises: constructing a target prediction function of the front wheel steering angle, the lateral distance deviation and the yaw angle deviation based on a preset prediction control equation, and obtaining a first constraint condition corresponding to the front wheel steering angle, a second constraint condition corresponding to the lateral distance deviation and a third constraint condition corresponding to the yaw angle deviation according to the target prediction function; based on the first constraint condition, the second constraint condition and the third constraint condition, determining an optimal path control sequence of the vehicle to obtain a preset prediction model of the vehicle, and tracking the trajectory planning path based on the preset prediction model.

5. The method of claim 1, wherein, The method of obtaining the steering wheel torque of the vehicle comprises: obtaining a driver's hand force, a driving speed of the vehicle and a steering wheel turning angle; based on a preset torque mapping relationship, generating a hand force torque of the driver's hand force, a damping torque of the driving speed of the vehicle and a return torque of the steering wheel turning angle; performing torque addition on the hand force torque, the damping torque and the return torque to obtain the steering wheel torque of the vehicle.

6. The method of claim 1, wherein, The preset exponential decay expectation function is: e y (k+i | k) = (1 - e -λi )e y (k) wherein e y is a lateral distance deviation of a center position of the vehicle from a center position of a road on which the vehicle is located, (1 - e -λi ) is a desired trajectory coefficient, k is a current period, i is a number of prediction periods, (k+i | k) is a state of a (k+i)th period predicted in a kth period, e -λi is an exponential decay function coefficient, e y (k) is a first value of the lateral distance deviation in the kth period.

7. A model prediction based lane keeping device, characterized by, The method comprises: The acquisition module is configured to acquire a front wheel steering angle of the vehicle, a lateral distance deviation between a center position of the vehicle and a center position of a road where the vehicle is located, and a yaw angle deviation of the vehicle, and construct a target lateral dynamics model according to the front wheel steering angle, the lateral distance deviation, and the yaw angle deviation. The path tracking module is configured to optimize the lateral distance deviation and the yaw angle deviation by using a preset exponential decay expectation function based on the target lateral dynamics model, to obtain a trajectory planning path of the vehicle, and track the trajectory planning path by using a preset prediction model to obtain a lane keeping request torque of the vehicle. The control module is configured to acquire a steering wheel torque of the vehicle, obtain a target torque of the vehicle according to the steering wheel torque and the lane keeping request torque, and control the vehicle to keep the lane according to the target torque.

8. The apparatus of claim 7, wherein, Before acquiring the front wheel steering angle of the vehicle, the lateral distance deviation between the center position of the vehicle and the center position of the road where the vehicle is located, and the yaw angle deviation of the vehicle, the acquisition module further comprises: An acquisition unit is configured to acquire driving information of the vehicle, steering wheel input information, and road information where the vehicle is located. A data cleaning unit is configured to perform data cleaning on the driving information, the steering wheel input information, and the road information where the vehicle is located based on a preset data cleaning strategy, to obtain cleaned driving information, steering wheel input information, and road information.

9. A vehicle characterized by comprising: The computer program is stored in the memory and executable on the processor, and the processor executes the program to implement the model prediction based lane keeping method according to any one of claims 1-6. The program is executed by the processor to implement the model prediction based lane keeping method according to any one of claims 1-6.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, ​

Images