Vehicle transverse control method and device, storage medium and electronic equipment

Abstract

The invention provides a vehicle transverse control method and device, a storage medium and electronic equipment, and relates to the field of smart mines, automatic driving and unmanned vehicles. The vehicle transverse control method comprises the steps of obtaining a target dynamic model constructed based on a side slip angle and transverse dynamic characteristics of a vehicle steering wheel; calculating an actual turning angle value of a steering wheel of the target vehicle by utilizing the target kinetic model; and adjusting the current turning angle value of the steering wheel of the target vehicle based on the actual turning angle value of the steering wheel of the target vehicle so as to realize transverse control of the target vehicle. According to the invention, the difficulty of a vehicle transverse control algorithm in engineering application is reduced. Meanwhile, the scheme does not depend on a complex vehicle model and various real-time state information, the requirement for the precision of the vehicle model is lowered, the complexity of a vehicle transverse control algorithm is simplified, and the vehicle transverse control algorithm is easier to implement.

Description

Technical Field

[0001] The present application relates to the technical fields of unmanned driving, automatic driving and unmanned vehicles, and specifically to a vehicle lateral control method and device, a storage medium and an electronic device. Background Art

[0002] At present, the research on vehicle lateral control algorithms has made some progress, but these algorithms are usually more complex, have high requirements on the accuracy of vehicle models, and need to obtain multiple status information of the vehicle in real time. In addition, in the application of unmanned mining vehicles, vehicles of different models, batches and ages often have poor consistency of wire control systems. The above factors also increase the difficulty of engineering application of vehicle lateral control algorithms. Summary of the invention

[0003] In view of this, embodiments of the present application provide a vehicle lateral control method and device, a storage medium, and an electronic device.

[0004] In a first aspect, an embodiment of the present application provides a vehicle lateral control method, including: obtaining a target dynamics model constructed based on the sideslip angle and lateral dynamic characteristics of the vehicle's steering wheel; using the target dynamics model, calculating the actual steering angle value of the target vehicle's steering wheel; based on the actual steering angle value of the target vehicle's steering wheel, adjusting the current steering angle value of the target vehicle's steering wheel to achieve lateral control of the target vehicle.

[0005] In combination with the first aspect, in certain implementations of the first aspect, the target dynamics model defines a mathematical relationship between an actual steering angle value of a vehicle steering wheel and a corrected theoretical steering angle value of the vehicle steering wheel; preferably, the mathematical relationship is expressed in the form of a polynomial.

[0006] In combination with the first aspect, in certain implementations of the first aspect, the polynomial includes static parameters and polynomial coefficients; wherein the static parameters are used to characterize the inherent characteristics of the vehicle, and the static parameters are determined during the vehicle calibration process and remain unchanged during the vehicle's service life; the polynomial coefficients change with changes in the vehicle's steering mechanism.

[0007] In combination with the first aspect, in certain implementations of the first aspect, the theoretical steering angle value of the vehicle's steering wheel is obtained by mathematical calculation using the vehicle's wheelbase and expected driving curvature as input parameters of the basic dynamics model; the corrected theoretical steering angle value of the vehicle's steering wheel is obtained by mathematical calculation using the vehicle's wheelbase, expected driving curvature and vehicle speed parameters as input parameters of the basic dynamics model; preferably, the vehicle speed parameters include the mth power of the longitudinal vehicle speed and / or the mth power of the characteristic vehicle speed, wherein the characteristic vehicle speed represents the speed characteristics of the vehicle under a specific driving scenario; preferably, the expected driving curvature represents the curvature between the center point of the driven wheel of the vehicle and the target aiming point on the expected driving path of the vehicle.

[0008] In combination with the first aspect, in certain implementations of the first aspect, the target dynamics model is used to calculate the actual steering angle value of the target vehicle's steering wheel, including: using a recursive least squares algorithm to automatically learn and determine the estimated value of each coefficient in the polynomial; based on the estimated value of each coefficient in the polynomial and the corrected theoretical steering angle value of the target vehicle's steering wheel, calculating the actual steering angle value of the target vehicle's steering wheel.

[0009] In combination with the first aspect, in certain implementations of the first aspect, a recursive least squares algorithm is used to automatically learn and determine an estimated value of each coefficient in a polynomial, including: in the Nth control cycle of the target vehicle, obtaining the estimated values ​​of each coefficient in the gain matrix updated in the N-1th control cycle and the polynomial updated in the N-1th control cycle, where N is a positive integer greater than or equal to 2; constructing a regression vector containing parameters to be identified of different powers; calculating the gain vector based on the gain matrix and the regression vector updated in the N-1th control cycle; updating the estimated value of each coefficient in the polynomial in the Nth control cycle based on the gain vector and the prediction error in the N-1th control cycle, and the estimated value of each coefficient in the polynomial updated in the N-1th control cycle; preferably, the parameter to be identified is the corrected theoretical steering angle value of the steering wheel of the target vehicle; preferably, the method also includes: updating the gain matrix in the Nth control cycle based on the gain vector and the regression vector, and the gain matrix updated in the N-1th control cycle.

[0010] In combination with the first aspect, in certain implementations of the first aspect, in the first control cycle, the gain matrix is ​​initialized as a diagonal matrix, and the estimated value of each coefficient in the initialization polynomial is zero; preferably, the value of the elements on the diagonal of the diagonal matrix is ​​1; preferably, the method also includes: after obtaining the estimated value of each coefficient in the polynomial in the updated first control cycle, updating the gain matrix in the first control cycle.

[0011] In a second aspect, an embodiment of the present application provides a vehicle lateral control device, including: an acquisition module for acquiring a target dynamics model constructed based on the sideslip angle and lateral dynamic characteristics of the vehicle's steering wheel; a calculation module for calculating the actual steering angle value of the target vehicle's steering wheel using the target dynamics model; and a control module for adjusting the current steering angle value of the target vehicle's steering wheel based on the actual steering angle value of the target vehicle's steering wheel to achieve lateral control of the target vehicle.

[0012] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program for executing the vehicle lateral control method described in the first aspect.

[0013] In a fourth aspect, an embodiment of the present application provides an electronic device, which includes: a processor; a memory for storing instructions executable by the processor; and the processor is used to execute the vehicle lateral control method described in the first aspect.

[0014] In the present application, by constructing a target dynamics model based on the sideslip angle and lateral dynamic characteristics of the vehicle's steering wheel, the actual steering angle value of the target vehicle's steering wheel is accurately calculated, and the current steering angle value of the target vehicle's steering wheel is adjusted accordingly to achieve lateral control of the target vehicle. This solution can automatically compensate for the consistency differences of the wire control system for vehicles of different models, batches and ages by adjusting the current steering angle value of the target vehicle's steering wheel, thereby reducing the difficulty of the vehicle lateral control algorithm of the present application in engineering applications. At the same time, the solution does not rely on complex vehicle models and multiple real-time status information, reduces the requirements for vehicle model accuracy, simplifies the complexity of the vehicle lateral control algorithm, and makes the vehicle control algorithm of the present application easier to implement. In addition, the lateral parameters of the target vehicle's steering wheel can be updated in a timely manner under different working environments of the target vehicle, so that the vehicle lateral control method of the present application has higher adaptability and generalization performance. Finally, the sideslip angle of the vehicle's steering wheel is incorporated into the target dynamics model to solve the problem. Figure 1 The limitations of the vehicle kinematic model shown can more accurately predict the response of the target vehicle under a given steering input, thereby improving the accuracy and stability of the lateral control of the target vehicle and enhancing the overall performance and safety of the target vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 Shown is a schematic diagram of a vehicle kinematic model in a global coordinate system provided by an embodiment of the present application.

[0017] Figure 2 Shown is a flow chart of a vehicle lateral control method provided in one embodiment of the present application.

[0018] Figure 3 The figure is a schematic diagram of a process for calculating the actual turning angle value of the steering wheel of a target vehicle provided by an embodiment of the present application.

[0019] Figure 4 The figure is a schematic diagram of a flow chart of determining estimated values ​​of coefficients in a polynomial provided by an embodiment of the present application.

[0020] Figure 5 Shown is a schematic structural diagram of a vehicle lateral control device provided in one embodiment of the present application.

[0021] Figure 6 Shown is a schematic structural diagram of an electronic device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] Figure 1 The figure shows a schematic diagram of a vehicle kinematic model in a global coordinate system provided by an embodiment of the present application. In this model, a four-wheeled vehicle is simplified into a front and rear two-wheeled form, assuming that P1 is a driven wheel and P2 is a steering wheel, the wheels have no lateral slip, and only the steering wheel can be steered. If the vehicle kinematic model is restricted to movement in a plane, the constraint equations of the driven wheel P1 and the steering wheel P2 can be expressed as:

[0024] xfsin(θ+δr)-yfcos(θ+δr)=0

[0025] xrsin(θ)-yrcos(θ)=0

[0026] Among them, (xr, yr) is the global coordinate of the center point of the driven wheel P1, (xf, yf) is the global coordinate of the center point of the steering wheel P2, θ is the driving direction of the vehicle in the global coordinate system, and θ r It is the actual steering angle value of the vehicle steering wheel in the vehicle body coordinate system.

[0027] Furthermore, if the wheelbase of the vehicle is L and the turning radius is R, the curvature of the vehicle is ρ = 1 / R. δd represents the ideal turning angle of the vehicle steering wheel in the vehicle body coordinate system. Without considering the side slip angle of the wheel, it is assumed that δd = δr. Then, the mathematical relationship between the actual turning angle of the vehicle steering wheel and the curvature of the vehicle (i.e., the vehicle kinematic model) is:

[0028]

[0029] In practical applications, if the running curvature ρ of the target vehicle is known, the actual steering angle value of the steering wheel of the target vehicle can be directly calculated by the above formula, and then the target vehicle can be controlled to follow the path.

[0030] However, when the target vehicle is traveling at a high speed, or the theoretical steering angle of the target vehicle's steering wheel is large, there will be a large deviation between the actual steering angle of the target vehicle's steering wheel calculated using the above formula and the actual steering ability of the target vehicle's steering wheel, which will affect the effect of the vehicle's lateral control.

[0031] In order to reduce the above deviations, traditional methods usually compensate the calculated actual steering angle value of the target vehicle's steering wheel when using the above vehicle kinematic model, thereby achieving a better control effect, but the final control effect will still be affected by the compensation strategy.

[0032] Based on the above, this application is proposed. Figure 2 FIG. 1 is a flow chart of a vehicle lateral control method provided by an embodiment of the present application. Figure 1 As shown, in this embodiment, the method includes the following steps.

[0033] Step S210, obtaining a target dynamics model constructed based on the sideslip angle and lateral dynamics characteristics of the steering wheel of the vehicle.

[0034] Specifically, the sideslip angle of a vehicle's steering wheel refers to the angle at which the vehicle's steering wheel deviates from the vehicle's theoretical direction of travel, and is used to describe the degree of inclination of the vehicle's steering wheel relative to the longitudinal axis of the vehicle. In other words, when the vehicle turns while driving, the vehicle's steering wheel will deflect based on the steering angle contained in the driver's operation or the command of the automatic control system. This deflection angle is the sideslip angle. The sideslip angle of the vehicle's steering wheel is directly related to the lateral force and steering response of the vehicle when turning, and therefore has an important impact on the vehicle's lateral stability and handling. For example, refer to Figure 1 As shown, (δr-δd) is the side slip angle of the steering wheel of the vehicle. Exemplarily, the front wheels of the vehicle are steering wheels, and the rear wheels are driven wheels.

[0035] Lateral dynamic characteristics refer to the dynamic response characteristics of the vehicle in the lateral direction, which determines the performance of the vehicle when performing lateral movement (such as turning or avoiding obstacles). Therefore, good lateral dynamic characteristics help improve the vehicle's handling and stability.

[0036] The target dynamics model is a mathematical model that includes the lateral deviation characteristics of the vehicle's steering wheel, the vehicle's driving parameters, etc., and is used to describe and predict the dynamic behavior and response of the vehicle during driving. Optionally, the target dynamics model can simulate the movement of the vehicle under different inputs (such as steering wheel angle, accelerator pedal position and braking force) and different road conditions (such as road friction coefficient, slope and curvature), and can be used to predict the actual steering angle value of the vehicle's steering wheel.

[0037] Step S220, using the target dynamics model, calculate the actual steering angle value of the steering wheel of the target vehicle.

[0038] Exemplarily, the driving parameters of the target vehicle, such as vehicle speed, steering wheel angle, road friction coefficient, etc., are input into the target dynamics model, and the lateral force and yaw motion of the target vehicle are calculated by formulas, thereby obtaining the actual steering angle value of the steering wheel of the target vehicle.

[0039] Step S230: adjusting the current steering angle value of the steering wheel of the target vehicle based on the actual steering angle value of the steering wheel of the target vehicle.

[0040] In some embodiments, the current direction of the target vehicle is continuously monitored, and by comparing the current direction of the target vehicle with the desired direction, it is determined whether the target vehicle needs to be adjusted laterally. If the target vehicle deviates from the desired direction, the actual steering angle value of the steering wheel of the target vehicle is calculated to correct the current direction of the target vehicle, thereby achieving lateral control of the target vehicle.

[0041] In this embodiment, by constructing a target dynamics model based on the sideslip angle and lateral dynamic characteristics of the vehicle's steering wheel, the actual steering angle value of the target vehicle's steering wheel is accurately calculated, and the current steering angle value of the target vehicle's steering wheel is adjusted accordingly to achieve lateral control of the target vehicle. This solution can automatically compensate for the consistency differences of the wire control system for vehicles of different models, batches and ages by adjusting the current steering angle value of the target vehicle's steering wheel, thereby reducing the difficulty of the vehicle lateral control algorithm of the present application in engineering applications. At the same time, this solution does not rely on complex vehicle models and multiple real-time status information, reduces the requirements for vehicle model accuracy, simplifies the complexity of the vehicle lateral control algorithm, and makes the vehicle control algorithm of the present application easier to implement. In addition, the lateral parameters of the target vehicle's steering wheel can be updated in a timely manner under different working environments of the target vehicle, so that the vehicle lateral control method of the present application has higher adaptability and generalization performance. Finally, the sideslip angle of the vehicle's steering wheel is incorporated into the target dynamics model to solve the problem. Figure 1 The limitations of the vehicle kinematic model shown can more accurately predict the response of the target vehicle under a given steering input, thereby improving the accuracy and stability of the lateral control of the target vehicle and enhancing the overall performance and safety of the target vehicle.

[0042] In some embodiments of the present application, the target dynamics model defines a mathematical relationship between an actual steering angle value of a vehicle steering wheel and a corrected theoretical steering angle value of the vehicle steering wheel.

[0043] It is understandable that the theoretical turning angle value of the vehicle steering wheel is derived based on a simple kinematic model, which may not take into account the interaction between the tire and the road surface, the dynamic response of the vehicle, and other external factors that may affect the vehicle's driving. However, in actual driving, the behavior of the vehicle will be affected by a variety of complex factors, such as the side slip characteristics of the tire, the mass and inertia of the vehicle, the friction coefficient of the road surface, etc. Therefore, in this embodiment, in order to make the actual turning angle value of the vehicle steering wheel closer to the expected value, the theoretical turning angle value of the vehicle steering wheel needs to be corrected to compensate for the influence of the above factors. Exemplarily, the corrected theoretical turning angle value of the vehicle steering wheel is obtained by adjusting and optimizing the vehicle dynamics model, which can more accurately reflect the actual steering behavior of the vehicle under specific driving conditions.

[0044] In some embodiments, the target dynamics model establishes a mathematical framework for describing the relationship between the actual steering angle value of the vehicle steering wheel and the corrected theoretical steering angle value of the vehicle steering wheel. This relationship is expressed by a mathematical formula, so that the target dynamics model can calculate the actual steering angle value of the target vehicle steering wheel according to the actual driving parameters of the target vehicle during the application stage to achieve more accurate vehicle lateral control.

[0045] In some embodiments, the mathematical relationship is expressed in the form of a polynomial, that is, the relationship between the actual steering angle value of the vehicle steering wheel and the corrected theoretical steering angle value of the vehicle steering wheel is defined by a polynomial equation. Exemplarily, the polynomial form of the target dynamics model can be expressed as: Among them, δ dd Indicates the corrected theoretical steering angle value of the vehicle's steering wheel.

[0046] It should be noted that the target dynamics model in this embodiment is a quartic polynomial for example only. When constructing the target dynamics model, those skilled in the art may select an appropriate polynomial degree according to factors such as the vehicle usage scenario and the vehicle control accuracy.

[0047] Different from directly constructing the relationship between the actual turning angle value of the vehicle steering wheel and the theoretical turning angle value of the vehicle steering wheel, this embodiment constructs the relationship between the actual turning angle value of the vehicle steering wheel and the corrected theoretical turning angle value of the vehicle steering wheel, which can more accurately reflect the behavior of the vehicle under actual driving conditions, thereby improving the calculation accuracy of the actual turning angle value of the vehicle steering wheel. In addition, the mathematical relationship in the form of a polynomial can flexibly express the nonlinear relationship between the actual turning angle value of the vehicle steering wheel and the corrected theoretical turning angle value of the vehicle steering wheel. This method not only improves the calculation accuracy of the actual turning angle value of the vehicle steering wheel, but also increases the adaptability and flexibility of the target dynamics model, so that it can better cope with different driving conditions and vehicle characteristics.

[0048] In some embodiments of the present application, the polynomial includes static parameters and polynomial coefficients. The static parameters are used to characterize the inherent characteristics of the vehicle, and the static parameters are determined during the vehicle calibration process and remain unchanged during the vehicle's service life; the polynomial coefficients change with changes in the vehicle's steering mechanism.

[0049] In this embodiment, after these static parameters are determined during the vehicle calibration process, they remain unchanged during the vehicle's use cycle. In this way, the vehicle only needs to be calibrated once to determine these static parameters, and there is no need to repeat the calibration during the use of the vehicle. This reduces the amount of parameters for vehicle calibration during lateral control of the vehicle, reduces the demand for computing power and memory resources, saves computing costs and time, and ensures the consistency of the vehicle's lateral control performance at different stages of use.

[0050] In this embodiment, the polynomial coefficients are not fixed, but will change according to the actual operating conditions of the vehicle steering mechanism. For example, the steering mechanism includes key components such as the steering gear, the steering motor, and the steering knuckle. The wear, aging, or maintenance status of these components will affect the steering performance of the vehicle. Therefore, the polynomial coefficients need to be adjusted according to the actual working state of the steering mechanism to ensure that the target dynamics model can accurately reflect the steering behavior of the vehicle under different usage conditions.

[0051] For example, the gears in the steering mechanism of the vehicle are worn due to long-term use, which causes the actual value of the steering ratio of the gear to be different from the designed value. In this case, the error caused by this wear can be compensated by adjusting the coefficients in the polynomial, thereby ensuring the accuracy of the actual angle value of the vehicle's steering wheel, thereby maintaining the steering accuracy and stability of the vehicle. In addition, changes in the steering mechanism also involve factors such as the aging of elastic elements in the steering system and changes in the viscosity of the lubricating oil, which will affect the steering accuracy of the vehicle. By adjusting the polynomial coefficients, the steering mechanism of the vehicle can adapt to these changes. This design makes the target dynamics model more adaptable and flexible, and can continuously provide accurate lateral control of the vehicle, and ensure the effectiveness of the lateral control algorithm even when the steering mechanism changes.

[0052] In some embodiments of the present application, the theoretical turning angle value of the vehicle's steering wheel is obtained by mathematical calculation using the vehicle's wheelbase and expected driving curvature as input parameters of a basic dynamics model.

[0053] Continuing with the above example, the basic dynamics model is δd=arctan(Lρ). In this embodiment, the wheelbase of the vehicle is substituted as a known term into the parameter L in the basic dynamics model, and the expected running curvature is substituted as a known term into the parameter ρ in the basic dynamics model.

[0054] Optionally, the expected driving curvature represents the curvature between the center point of the driven wheel of the vehicle and the target aiming point on the expected driving path of the vehicle, describing the curvature of the vehicle's driving trajectory. Exemplarily, the target aiming point is an idealized reference point located on the expected driving path of the vehicle and is used to guide the steering operation of the vehicle to ensure that the vehicle can travel according to the predetermined trajectory.

[0055] Optionally, the corrected theoretical turning angle value of the vehicle steering wheel is obtained by mathematical calculation using the vehicle's wheelbase, expected driving curvature and vehicle speed parameters as input parameters of the basic dynamics model.

[0056] For example, the mathematical expression of the corrected theoretical steering angle of the vehicle steering wheel is δdd=arctan(LρV c ), where V c Indicates vehicle speed parameter.

[0057] It is understandable that when the vehicle speed changes, the dynamic response of the vehicle, especially the response of the steering wheel, will also change. For example, when driving at high speed, the steering of the vehicle will become more sensitive, and a small steering angle value of the vehicle steering wheel may cause a large yaw motion of the vehicle; while when driving at low speed, the steering of the vehicle is relatively blunt, and a larger steering angle value of the vehicle steering wheel is required to achieve the same yaw motion. Therefore, by incorporating the vehicle speed parameter into the calculation of the corrected theoretical steering angle value of the vehicle steering wheel, the corrected theoretical steering angle value of the vehicle steering wheel can be made closer to the actual driving situation, so that the vehicle can maintain good handling performance and stability.

[0058] Optionally, the vehicle speed parameter includes the mth power of the longitudinal vehicle speed and / or the mth power of the characteristic vehicle speed.

[0059] Exemplarily, based on this, the mathematical expression of the corrected theoretical steering angle value of the vehicle steering wheel is: Among them, V x Indicates the longitudinal speed, V ch Indicates characteristic vehicle speed.

[0060] The longitudinal speed refers to the linear speed of the vehicle along its forward direction. The characteristic speed refers to the speed characteristics of the vehicle in a specific driving scenario. In some embodiments, the characteristic speed refers to the speed corresponding to the maximum yaw rate gain in the steady-state response analysis of the vehicle. When the vehicle is driving at the characteristic speed and performing a steering operation, a small change in the steering angle of the vehicle steering wheel will result in a large yaw motion. At this time, the steering sensitivity of the vehicle reaches the highest at this characteristic speed.

[0061] Specifically, when driving at high speed, the aerodynamic effect, tire side slip characteristics, and dynamic response of the suspension system will become more complicated as the vehicle speed increases. For example, when driving at high speed, the air resistance of the vehicle will increase significantly as the vehicle speed increases. This relationship is not linear, and the m-th power of the longitudinal vehicle speed can capture the nonlinear effect of the longitudinal vehicle speed on the vehicle dynamics. Therefore, by introducing the m-th power of the longitudinal vehicle speed, this nonlinear characteristic can be simulated more accurately.

[0062] Similarly, by introducing the mth power of the characteristic vehicle speed, the dynamic response of the vehicle at different speed levels can be captured more carefully. This processing method is particularly suitable for nonlinear target dynamics models and can better simulate the complex behavior of vehicles in actual driving. In addition, the mth power of the characteristic vehicle speed can also take into account the interaction between the vehicle speed and other vehicle parameters (such as wheelbase and tire characteristics), further improving the prediction accuracy of the target dynamics model.

[0063] Optionally, the m in the mth power of the longitudinal vehicle speed and / or the mth power of the characteristic vehicle speed, and the characteristic vehicle speed are static parameters. In some embodiments, the static parameters and polynomial coefficients included in the term are parameter sets to be solved for the target dynamics model. The specific values ​​of the parameter set are related to the type of vehicle, the load of the vehicle, etc. For example, the actual values ​​of the parameter set of the target dynamics model under no-load and heavy-load conditions are shown in Table 1.

[0064] Table 1

[0065] Loading status <![CDATA[v ch ]]> m <![CDATA[α 0 ]]> <![CDATA[α 1 ]]> <![CDATA[α 2 ]]> <![CDATA[α 3 ]]> <![CDATA[α 4 ]]> No load 10.833 3 0.0 1.036 0.1744 -0.0045 -0.5580 Overload 12.778 3 0.0 1.2175 0.0731 -0.1883 -0.1885

[0066] Figure 3 FIG. 1 is a flow chart of calculating the actual turning angle value of the steering wheel of the target vehicle provided by an embodiment of the present application. Figure 3 As shown, the actual turning angle value of the steering wheel of the target vehicle is calculated by using the target dynamics model, including the following steps.

[0067] Step S310: automatically learning and determining the estimated value of each coefficient in the polynomial using a recursive least squares algorithm.

[0068] The recursive least squares algorithm is an adaptive filtering algorithm that can recursively update the best estimate of the parameter set of the target dynamics model when new data (in this application, the new data refers to the wheelbase of the vehicle, the driving curvature of the vehicle, the longitudinal speed, etc.) arrives. Automatic learning means that the recursive least squares algorithm can automatically learn from the new data and update the parameter set of the target dynamics model in real time without human intervention.

[0069] In addition, it should be noted that the static parameters in the parameter set of the target dynamic model usually remain unchanged after calibration. Therefore, this embodiment uses a recursive least squares algorithm to automatically learn and determine the estimated values ​​of the polynomial coefficients in the parameter set.

[0070] Step S320, calculating the actual steering angle value of the steering wheel of the target vehicle based on the estimated value of each coefficient in the polynomial and the corrected theoretical steering angle value of the steering wheel of the target vehicle.

[0071] Exemplarily, the target dynamics model is a quadratic polynomial model, i.e., δr = a 0 +a 1 δ dd +a 2 In step S310, the recursive least squares algorithm has automatically learned and updated the estimated values ​​of these polynomial coefficients according to the new data. Assume that the estimated values ​​of the polynomial coefficients are: 0 =0.1, a 1 =0.9, a 2=0.05. Assuming that the corrected theoretical steering angle of the target vehicle's steering wheel is 3°, then the actual steering angle of the target vehicle's steering wheel = 0.1 + 0.9 × 3° + 0.05 × 3 2 =3.25°.

[0072] In this embodiment, the recursive least squares algorithm can recursively update the estimated values ​​of the polynomial coefficients using the new data, because the recursive least squares algorithm does not need to reprocess the entire data set (including the new data and the historical data corresponding to the new data), so the computational complexity can be reduced, and this method is particularly suitable for dynamically changing scenes. In addition, the actual turning angle value of the steering wheel of the target vehicle is calculated using the estimated values ​​of these optimized polynomial coefficients, so that the vehicle lateral control strategy can accurately adapt to the real-time needs of the target vehicle, thereby enhancing driving safety.

[0073] Figure 4 FIG. 1 is a flow chart of determining the estimated values ​​of each coefficient in a polynomial according to an embodiment of the present application. Figure 3 Based on the embodiment shown, Figure 4 The embodiment shown is described below in detail. Figure 4 The embodiment shown and Figure 3 The differences and similarities of the illustrated embodiments are not described in detail.

[0074] like Figure 4 As shown, in this embodiment, the recursive least squares algorithm is used to automatically learn and determine the estimated value of each coefficient in the polynomial, including the following steps.

[0075] Step S410, in the Nth control cycle of the target vehicle, obtaining the estimated values ​​of the coefficients in the gain matrix updated in the N-1th control cycle and the polynomial updated in the N-1th control cycle.

[0076] The control cycle refers to the time interval for the vehicle control system to make a complete control decision and execution. Exemplarily, the control cycle includes "perception, decision, execution and feedback". The gain matrix represents the covariance of the estimated values ​​of the polynomial coefficients and is a measure of the error of the estimated values ​​of the polynomial coefficients. The gain matrix is ​​updated in the N-1th control cycle to reflect the uncertainty changes in the estimated values ​​of the polynomial coefficients after each update. Exemplarily, the gain matrix updated in the N-1th control cycle is denoted as P n-1 .

[0077] It should be noted that the size of the gain matrix is ​​related to the number of polynomial coefficients. For example, the number of polynomial coefficients is a 0 、a 1 、a 2 、a 3 、a 4, then the size of the gain matrix is ​​5×5. Below, for the convenience of description, the target dynamics model is Take this as an example to explain this embodiment in detail.

[0078] Step S420, constructing a regression vector containing parameters to be identified of different powers.

[0079] In some embodiments, the parameter to be identified is the corrected theoretical steering angle value of the steering wheel of the target vehicle. Then, let x = δdd.

[0080] As mentioned above, regression vectors of the parameters to be identified with different powers can fit nonlinear patterns in the data. By including different powers of the parameters to be identified, the target dynamics model can adapt to complex changes in the data. For example, if the relationship in the data is curvilinear, it cannot be accurately fitted using a linear model, and regression vectors including higher powers can provide a better fit. In addition, the target dynamics model in polynomial form can adapt to various data distributions by adjusting the coefficients of the powers, and therefore has better generalization ability. Finally, it should be noted that the data here refers to the wheelbase of the vehicle, the driving curvature of the vehicle, the longitudinal speed, etc.

[0081] Step S430, calculating a gain vector based on the gain matrix and regression vector updated in the N-1th control cycle.

[0082] The gain vector is a weight vector that indicates the importance of new data in updating the estimated values ​​of the polynomial coefficients. Among them, 1 is the forgetting factor, which means that all historical data are given the same weight, that is, no historical data is forgotten. It should be noted that the historical data here corresponds to the new data, that is, the historical data includes the previously determined vehicle wheelbase, vehicle running curvature, longitudinal speed, etc.

[0083] Step S440, based on the gain vector and the prediction error in the N-1th control cycle, and the estimated value of each coefficient in the polynomial updated in the N-1th control cycle, update the estimated value of each coefficient in the polynomial in the N-1th control cycle.

[0084] For example, let Pa be the polynomial coefficient, that is, Pa = [a 0 a 1 a 2 a 3 a 4 ] T .

[0085] The estimated values ​​of the polynomial coefficients in the Nth control cycle can be expressed as in, Represents the prediction error within the N-1th control cycle.

[0086] In this embodiment, in the Nth control cycle, the estimated values ​​of the gain matrix and polynomial coefficients updated in the N-1th control cycle are first obtained, wherein the gain matrix determines the degree of influence of the new data on the estimated values ​​of the polynomial coefficients. Next, a regression vector containing parameters to be identified of different powers is constructed, and this regression vector contains the corrected theoretical turning angle value of the steering wheel of the target vehicle and its higher-order terms to capture more complex nonlinear relationships. Then, the gain vector is calculated based on the gain matrix and regression vector updated in the N-1th control cycle, wherein the gain vector indicates the correction direction and amplitude of the estimated values ​​of the current polynomial coefficients by the new data. Then, the estimated values ​​of the polynomial coefficients in the Nth control cycle are updated using the gain vector and the prediction error in the N-1th control cycle. This updating process minimizes the prediction error, thereby improving the prediction accuracy of the model. This embodiment designs a concise and generalized self-learning mechanism that can identify the estimated values ​​of the polynomial coefficients in real time. This improvement in adaptive capability avoids the problem of poor consistency of the wire control system due to different vehicle batches and vehicle ages, and reduces the resource consumption of manual parameter adjustment.

[0087] In some embodiments of the present application, after executing step S440, the gain matrix in the Nth control cycle is further updated based on the gain vector and the regression vector, and the gain matrix updated in the N-1th control cycle.

[0088] For example, the gain matrix in the Nth control cycle is This updating process ensures that the gain matrix can adapt to changes in vehicle dynamics, thereby providing more accurate parameter estimates in each control cycle. In this way, the vehicle's steering control system can dynamically adapt to various driving conditions, ensuring that the target vehicle can maintain optimal steering performance and stability in different situations.

[0089] In some embodiments of the present application, in the first control cycle, the gain matrix is ​​initialized to a diagonal matrix, and the estimated value of each coefficient in the initialization polynomial is zero.

[0090] Specifically, the diagonal matrix has the advantages of simple structure and high computational efficiency, which gives the recursive least squares algorithm a clear and easy-to-handle starting point before starting iteration. Furthermore, the estimated values ​​of the polynomial coefficients are initialized to zero, which simplifies the setting of the initial state and does not cause deviation to the estimated values ​​of the polynomial coefficients when the recursive least squares algorithm starts iterating.

[0091] With this initialization, the recursive least squares algorithm can start from a simple and computationally efficient state and gradually adjust and optimize the estimates of the gain matrix and polynomial coefficients as new data arrives. This approach helps ensure the stability and convergence of the recursive least squares algorithm, while allowing the recursive least squares algorithm to quickly adapt to the actual vehicle dynamics after collecting more new data, and to effectively adjust parameters in subsequent control cycles to achieve accurate vehicle lateral control.

[0092] Optionally, the element values ​​on the diagonal of the diagonal matrix are 1. This means that in the initial state, the weights of all parameters are equal, which helps to prevent the recursive least squares algorithm from being too sensitive to certain parameters and ignoring other parameters in the early stage, thereby ensuring the fairness and balance of the recursive least squares algorithm. Secondly, this setting simplifies the implementation of the recursive least squares algorithm, because in the initial state, the update of each parameter depends only on its own error and is not affected by other parameters, which allows the recursive least squares algorithm to respond quickly to new data at the beginning without the need for complex calculations to adjust the relative importance of different parameters.

[0093] Optionally, the method further includes: after obtaining the estimated values ​​of the coefficients in the polynomial in the updated first control period, updating the gain matrix in the first control period. The specific updating method can be found in the description of the above embodiment, which will not be repeated here.

[0094] Combination of the above Figures 1 to 4 , describes in detail the vehicle lateral control method embodiment of the present application, and the following is combined with Figure 5 , the vehicle lateral control device embodiment of the present application is described in detail. It should be understood that the description of the vehicle lateral control method embodiment corresponds to the description of the vehicle lateral control device embodiment, so the parts not described in detail can refer to the previous method embodiment.

[0095] Figure 5 FIG. 1 is a schematic diagram of the structure of a vehicle lateral control device provided by an embodiment of the present application. Figure 5 As shown, the vehicle lateral control device 50 provided in the embodiment of the present application includes:

[0096] An acquisition module 510 is used to acquire a target dynamics model constructed based on the sideslip angle and lateral dynamic characteristics of the steering wheel of the vehicle;

[0097] A calculation module 520 is used to calculate the actual turning angle value of the steering wheel of the target vehicle using the target dynamics model;

[0098] The control module 530 is used to adjust the current steering angle value of the steering wheel of the target vehicle based on the actual steering angle value of the steering wheel to achieve lateral control of the target vehicle.

[0099] In one embodiment of the present application, the target dynamics model defines a mathematical relationship between an actual steering angle value of a steering wheel of a vehicle and a corrected theoretical steering angle value of a steering wheel of the vehicle; preferably, the mathematical relationship is expressed in the form of a polynomial.

[0100] In one embodiment of the present application, the polynomial includes static parameters and polynomial coefficients; wherein the static parameters are used to characterize the inherent characteristics of the vehicle, and the static parameters are determined during the vehicle calibration process and remain unchanged during the vehicle's service life; the polynomial coefficients change with changes in the vehicle's steering mechanism.

[0101] In one embodiment of the present application, the theoretical turning angle value of the steering wheel of the vehicle is obtained by mathematical calculation using the wheelbase and expected running curvature of the vehicle as input parameters of the basic dynamics model;

[0102] The corrected theoretical steering angle of the steering wheel of the vehicle is obtained by mathematical calculation using the vehicle's wheelbase, expected running curvature and vehicle speed parameters as input parameters of the basic dynamics model;

[0103] Preferably, the vehicle speed parameter includes the mth power of the longitudinal vehicle speed and / or the mth power of the characteristic vehicle speed, wherein the characteristic vehicle speed represents the speed characteristic of the vehicle in a specific driving scenario;

[0104] Preferably, the expected driving curvature represents the curvature between the center point of the driven wheels of the vehicle and the target aiming point on the expected driving path of the vehicle.

[0105] In one embodiment of the present application, the calculation module 520 is also used to automatically learn and determine the estimated value of each coefficient in the polynomial using a recursive least squares algorithm; based on the estimated value of each coefficient in the polynomial and the corrected theoretical steering angle value of the steering wheel of the target vehicle, calculate the actual steering angle value of the steering wheel of the target vehicle.

[0106] In one embodiment of the present application, the calculation module 520 is also used to obtain, in the Nth control cycle of the target vehicle, the estimated values ​​of each coefficient in the gain matrix updated in the N-1th control cycle and the polynomial updated in the N-1th control cycle, where N is a positive integer greater than or equal to 2; construct a regression vector containing parameters to be identified of different powers; calculate the gain vector based on the gain matrix updated in the N-1th control cycle and the regression vector; update the estimated values ​​of each coefficient in the polynomial in the Nth control cycle based on the gain vector and the prediction error in the N-1th control cycle, and the estimated values ​​of each coefficient in the polynomial updated in the N-1th control cycle; preferably, the parameter to be identified is the corrected theoretical steering angle value of the steering wheel of the target vehicle; preferably, the calculation module 520 is also used to update the gain matrix in the Nth control cycle based on the gain vector and the regression vector, as well as the gain matrix updated in the N-1th control cycle.

[0107] In one embodiment of the present application, in the first control cycle, the gain matrix is ​​initialized as a diagonal matrix, and the estimated value of each coefficient in the initialization polynomial is zero; preferably, the value of the elements on the diagonal of the diagonal matrix is ​​1; preferably, the calculation module 520 is also used to update the gain matrix in the first control cycle after obtaining the estimated value of each coefficient in the polynomial in the updated first control cycle.

[0108] Below, reference Figure 6 To describe an electronic device according to an embodiment of the present application. Figure 6 Shown is a schematic structural diagram of an electronic device provided by an exemplary embodiment of the present application.

[0109] like Figure 6 As shown, the electronic device 60 includes one or more processors 601 and a memory 602 .

[0110] The processor 601 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 60 to perform desired functions.

[0111] The memory 602 may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 601 may run the program instructions to implement the vehicle lateral control method of each embodiment of the present application described above and / or other desired functions. Various contents such as the target dynamics model, the actual angle value of the steering wheel of the target vehicle, and the current angle value of the steering wheel of the target vehicle may also be stored in the computer-readable storage medium.

[0112] In one example, the electronic device 60 may further include: an input device 603 and an output device 604, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0113] The input device 603 may include, for example, a keyboard, a mouse, etc.

[0114] The output device 604 can output various information to the outside, including the target dynamics model, the actual turning angle value of the steering wheel of the target vehicle, the current turning angle value of the steering wheel of the target vehicle, etc. The output device 604 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0115] Of course, to simplify, Figure 6 Only some of the components related to the present application in the electronic device 60 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device 60 may also include any other appropriate components.

[0116] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the vehicle lateral control method according to various embodiments of the present application described above in this specification.

[0117] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0118] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the vehicle lateral control method according to various embodiments of the present application described above in this specification.

[0119] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0120] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0121] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0122] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0123] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0124] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A vehicle lateral control method, characterized in that: include: Obtain a target dynamics model based on the sideslip angle and lateral dynamic characteristics of the vehicle's steering wheel; Utilizing the target dynamics model, calculating the actual turning angle value of the steering wheel of the target vehicle; Based on the actual steering angle value of the steering wheel of the target vehicle, the current steering angle value of the steering wheel of the target vehicle is adjusted to achieve lateral control of the target vehicle.

2. The vehicle lateral control method according to claim 1, characterized in that: The target dynamics model defines a mathematical relationship between an actual turning angle value of the vehicle steering wheel and a corrected theoretical turning angle value of the vehicle steering wheel; Preferably, the mathematical relationship is expressed in the form of a polynomial.

3. The vehicle lateral control method according to claim 2, characterized in that: The polynomial includes static parameters and polynomial coefficients; wherein the static parameters are used to characterize the inherent characteristics of the vehicle, and the static parameters are determined during the vehicle calibration process and remain unchanged during the vehicle's service life; the polynomial coefficients change with changes in the vehicle's steering mechanism.

4. The vehicle lateral control method according to claim 2, characterized in that: The theoretical turning angle value of the steering wheel of the vehicle is obtained by mathematical calculation using the wheelbase and expected running curvature of the vehicle as input parameters of the basic dynamics model; The corrected theoretical turning angle value of the vehicle steering wheel is obtained by mathematical calculation using the vehicle wheelbase, the expected running curvature and the vehicle speed parameters as input parameters of the basic dynamics model; Preferably, the vehicle speed parameter includes the mth power of the longitudinal vehicle speed and / or the mth power of the characteristic vehicle speed, wherein the characteristic vehicle speed represents the speed characteristic of the vehicle in a specific driving scenario; Preferably, the expected driving curvature represents the curvature between the center point of the driven wheel of the vehicle and the target aiming point on the expected driving path of the vehicle.

5. The vehicle lateral control method according to any one of claims 2 to 4, characterized in that: The method of calculating the actual turning angle value of the steering wheel of the target vehicle by using the target dynamics model includes: Automatically learning and determining estimated values ​​of coefficients in the polynomial using a recursive least squares algorithm; Based on the estimated values ​​of the coefficients in the polynomial and the corrected theoretical steering angle value of the steering wheel of the target vehicle, the actual steering angle value of the steering wheel of the target vehicle is calculated.

6. The vehicle lateral control method according to claim 5, characterized in that: The method of automatically learning and determining the estimated values ​​of the coefficients in the polynomial using a recursive least squares algorithm includes: In the Nth control cycle of the target vehicle, obtaining the estimated values ​​of the coefficients of the gain matrix updated in the N-1th control cycle and the polynomial updated in the N-1th control cycle, where N is a positive integer greater than or equal to 2; Construct a regression vector containing the parameters to be identified at different powers; Calculating a gain vector based on the gain matrix updated in the N-1th control period and the regression vector; Based on the gain vector and the prediction error in the N-1th control period, and the estimated value of each coefficient in the polynomial updated in the N-1th control period, updating the estimated value of each coefficient in the polynomial in the N-1th control period; Preferably, the parameter to be identified is the corrected theoretical turning angle value of the steering wheel of the target vehicle; Preferably, the method further comprises: updating the gain matrix in the Nth control period based on the gain vector and the regression vector, and the gain matrix updated in the N-1th control period.

7. The vehicle lateral control method according to claim 6, characterized in that: In the first control cycle, the gain matrix is ​​initialized to a diagonal matrix, and the estimated value of each coefficient in the polynomial is initialized to zero; Preferably, the element values ​​on the diagonal of the diagonal matrix are 1; Preferably, the method further comprises: after obtaining the updated estimated value of each coefficient in the polynomial in the first control period, updating the gain matrix in the first control period.

8. A vehicle lateral control device, characterized in that: include: An acquisition module, used for acquiring a target dynamics model constructed based on the sideslip angle and lateral dynamics characteristics of the vehicle steering wheel; A calculation module, used to calculate the actual turning angle value of the steering wheel of the target vehicle using the target dynamics model; The control module is used to adjust the current steering angle value of the steering wheel of the target vehicle based on the actual steering angle value of the steering wheel of the target vehicle to achieve lateral control of the target vehicle.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and the computer program is used to execute the vehicle lateral control method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is used to execute the vehicle lateral control method described in any one of claims 1 to 7.

Images