Path tracking method and system for front-wheel steering and rear-wheel distributed driving vehicle

Abstract

The invention relates to the technical field of auxiliary driving, in particular to a front-wheel steering and rear-wheel distributed driving vehicle path tracking method. The method comprises the following steps: S1, establishing a nonlinear vehicle dynamics model of front-wheel steering and rear-wheel distributed driving; s2, establishing a path tracking task constraint; s3, designing an adaptive robust controller based on task constraints; s4, based on Lyapunov analysis, stability and constraint following error performance of the adaptive robust controller are analyzed; and S5, using the controller in the step S3 to control the vehicle to perform path tracking. According to the path tracking control algorithm provided by the invention, a model which better conforms to the actual dynamic characteristics of the vehicle is established, and the condition of relatively large initial tracking deviation is considered and processed, so that the control stability of the vehicle under the influence of uncertain factors and relatively large initial tracking deviation can be improved, and a tracking error sample is limited in a specified area.

Description

Technical Field

[0001] The present invention relates to the technical field of assisted driving, and in particular to a path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle. Background Art

[0002] Path tracking is one of the important functions of autonomous vehicles. Its purpose is to control the steering mechanism of the vehicle to make it follow a given desired path. The given desired path point information is generally issued by the decision-making planning module. The issued desired path point usually includes information such as the expected speed, lateral and longitudinal coordinates, and yaw angle. In the lateral path tracking of autonomous vehicles, it is necessary to design a reasonable controller so that the vehicle can complete the specified steering action to track the expected lateral and longitudinal coordinates and yaw angle.

[0003] Front-wheel steering and rear-wheel distributed drive vehicles have two control modes for vehicle steering: front-wheel steering control and rear axle active yaw moment control. For path tracking control methods with desired steering angle and active yaw moment as control outputs, there are currently methods such as sliding film control, model predictive control MPC, and composite nonlinear feedback control.

[0004] In the prior art, W. Zhenpo et al. proposed a hierarchical control architecture. The upper controller is based on a two-degree-of-freedom dynamics model and directly designs the sliding membrane surface with the tracking error as the variable. The designed sliding membrane control law outputs the angle and yaw moment control quantity to track the desired yaw rate. The lower controller designs a torque optimization distribution strategy to distribute the required torque to the drive wheels to enhance vehicle stability. However, the designed hierarchical sliding membrane control method focuses on achieving stability in vehicle control and steady-state performance of tracking error, while ignoring the changes in tracking error during transient convergence. Therefore, it cannot guarantee the transient and steady-state safety of the vehicle during path tracking. In addition, it directly uses the tracking error as the controlled object. When the initial tracking error is large, the control force based on error feedback is large and easily causes actuator saturation, lacking effective control under this special working condition.

[0005] Qifan Tan et al. proposed an upper-lower control architecture that combines model predictive control and torque distribution strategy. In the design of the upper-level controller, based on the path tracking error model, the general disturbance caused by the distribution error and the sudden disturbance caused by the external force are considered in the model. By converting the robustness constraints into linear matrix inequalities, the robust MPC control law output angle and expected yaw moment are obtained. The lower-level controller distributes the expected yaw moment obtained by the upper-level controller to the drive wheels, and then combines the steering control for path tracking. The hierarchical control architecture based on the model predictive control method also focuses on achieving the steady-state performance of the tracking error, aiming to eventually converge the tracking error to a very small interval, while ignoring the transient performance of the tracking error during the convergence process. In addition, its solution based on rolling optimization requires high computing power for the controller.

[0006] R. Wang et al. designed a composite nonlinear feedback controller consisting of a linear feedback part and a nonlinear feedback part. The linear part includes a reference yaw rate derivative term related to the change in road curvature. Considering that the vehicle lateral velocity is usually difficult to measure with low-cost sensors, a reduced-order observer is designed to estimate the lateral velocity using only the measured value of the yaw rate. When the vehicle tracking error is close to convergence to 0, the nonlinear part is used to increase the damping ratio to eliminate overshoot. This control method allows the vehicle yaw rate and lateral velocity to track their respective expected values ​​with relatively low overshoot and steady-state error. This method lacks consideration of the possible time-varying uncertainties in the actual vehicle dynamics. When the initial tracking error is large, the large error-based feedback term in the controller is prone to actuator saturation, and effective control of the vehicle cannot be guaranteed under large initial tracking error conditions.

[0007] In addition, the current commonly used lateral path tracking control method is difficult to deal with the actuator saturation problem that may be caused by a large initial tracking error, which will deteriorate the control performance. In the case of uncertainty in the vehicle model, this control instability caused by actuator saturation will be aggravated, the transient performance of the tracking error cannot be guaranteed, and the tracking error may be too large to exceed the safety boundary. Therefore, it is impossible to strictly guarantee the safety of the autonomous driving vehicle in performing the path tracking task under such initial working conditions. Summary of the invention

[0008] The purpose of the present invention is to propose a path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle in response to the problems existing in the background technology. The present invention designs a path tracking robust control method for front-wheel steering and rear-wheel distributed drive vehicles to ensure preset performance of tracking error. The control method mainly consists of vehicle dynamics modeling, tracking kinematics modeling, path tracking constraint establishment and robust controller design. The path tracking control algorithm proposed in the present invention establishes a model that is more in line with the actual dynamic characteristics of the vehicle, and takes into account and handles the situation of large initial tracking deviations. It can improve the control stability of the vehicle under the influence of uncertain factors and large initial tracking deviations, and limit the tracking error sample to a specified area.

[0009] A first aspect of the present invention provides a path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle, comprising the following specific steps: S1. Establish a nonlinear vehicle dynamics model with front-wheel steering and rear-wheel distributed drive; S11. Constructing a vehicle lateral dynamics model; S12, constructing a path tracking operation model; S2, establish path tracking task constraints; S21, introduce auxiliary variables to correct the bilateral constraints of lateral errors; S22, perform state bijective transformation to convert the bounded state into an unbounded state; S23, establishing path tracking equality constraints; S3. Design an adaptive robust controller based on task constraints; S31. Design a servo control law for the nominal system; S32, design the error feedback partial control law; S33. Design an adaptive compensation partial control law to compensate for the initial deviation and uncertainty of the system; S4. Stability and constrained following error performance of adaptive robust controller based on Lyapunov analysis; S5. Use the controller in step S3 to control the vehicle to perform path tracking.

[0010] Preferably, the nonlinear vehicle dynamics model of front-wheel steering and rear-wheel distributed drive established in step S1 includes: The vehicle longitudinal axis is Axis, the vehicle's horizontal axis direction is Axis, composition Body coordinate system, is the center of gravity of the vehicle; the center of gravity of the front wheelbase of the vehicle is , rear wheelbase center of gravity ; The front axle of the vehicle has a steering function. Assume that the steering angles of the two wheels on the front axle are consistent and set the front wheel steering angle Control interface; the two wheels on the rear axle are driving wheels, providing driving force / braking force , interface; The model also includes a positioning module, a perception module, a decision-making planning module and a control module; it is also equipped with a steering angle sensor and a wheel speed sensor to obtain the actual steering angle of the vehicle. and steering angular velocity ; The planning module provides the control module with the desired path information and the curvature information of the desired path The control module obtains the active yaw torque and the desired steering angle control amount according to the designed path tracking algorithm. The active yaw torque is input into the driving wheel and steering mechanism in the chassis module together with the steering angle after the driving force distribution.

[0011] Preferably, in step S11, a vehicle lateral dynamics model with a second-order inertia time lag is established as follows: in, is the vehicle mass, is the vehicle moment of inertia, and is a time-varying bounded uncertainty parameter caused by model errors and external disturbances; is the desired front wheel turning angle of the vehicle; and It is the system parameter of the time-delay second-order steering system, and its specific value needs to be obtained through steering test calibration; Other kinetic parameters ~ The specific expression is as follows: in, is the front wheel cornering stiffness, is the rear wheel cornering stiffness; in practice, the dynamic parameters in the above formula are It is difficult to identify, so we assume that the parameter ~ All by nominal part and the uncertain part which is unknown and changes with time composition.

[0012] Preferably, in step S12, the vehicle obtains its own longitudinal and transverse coordinates and yaw angle and other posture information in the geodetic coordinate system through the inertial measurement unit, and the vehicle obtains the longitudinal and transverse coordinates, yaw angle and road curvature information of the desired path point through the upper planning module; the yaw angle error of the vehicle relative to the desired path can be expressed as: in, is the heading angle error, is the yaw angle; is the ideal yaw angle at the desired path point. According to the road curvature, the ideal yaw angle change rate can be expressed as: is the longitudinal velocity; Then get The first-order and second-order forms of are as follows: The lateral error is defined as the shortest distance between the vehicle's current position and the nearest desired path point. According to the kinematic relationship, the first-order and second-order forms of the lateral error are as follows: To facilitate the subsequent controller design, the vehicle lateral dynamics formula is integrated and the state space equation based on error dynamics is established as follows: Among them, the vector is the state variable, For control input; other parts The specific expression is as follows: Preferably, in step S21, the path tracking lateral offset of the vehicle needs to meet the following boundary conditions: in, and are the minimum and maximum lateral offsets allowed, respectively; Introduction Definition 1: If the following conditions are met, it is a conversion function: (1) And in It is monotonically increasing on Keep on ,in is the preset time; (2) for And for all All have boundaries, is the order of the controlled system; The initial position of the vehicle may be far from the nearest point of the desired path, i.e. Too large; introduce auxiliary variables To correct the lateral error, the corrected error is as follows: in, , conversion function According to Definition 1, we select: Based on the above correction, the corrected lateral error has the following properties: The following constraints are imposed to correct the lateral error: The above formula can be restated as: Among them, the preset performance boundary parameters and The following inequality is satisfied: The original lateral error is restricted to a tighter band.

[0013] Preferably, in step S22, the bounded state Convert to unbounded state , the barrier transformation function based on the tangent function aims to achieve the following transformation: in, Derivative the above formula with respect to time t, we get: use To represent the vehicle yaw angle error, that is ,make , so the controlled state quantity Can be converted to a new state ; The following formula is obtained: in Since for all , both have Established, therefore is full rank; after the state transformation, about the unconstrained variables Status The kinetic equation can be expressed as: in because and are all full rank, so is also full rank, that is It is also reversible.

[0014] Preferably, in step S23, the goal of the path tracking task of the autonomous driving vehicle is to make the path tracking error converge to 0, that is, the vehicle travels on the desired path point; the convergence performance is converted into the following first-order and second-order equality constraints on the system state: Solving the differential equations above, the corrected lateral error and yaw angle error will be and , and approaches 0 over time; based on the characteristics of state transformation, the same constraints are imposed on the transformation state: Write the above equality constraint in matrix form , and order , which represents the constraint following error; in, , ; The second-order form of the constraints is: in, .

[0015] Preferably, in step S3, after the state transformation, the unconstrained variables The parameter matrix of the dynamic equation of state Z is decomposed into the following nominal part and uncertain part: The matrix with horizontal lines is the known nominal part, and the other matrices represent the unknown uncertain parts. To simplify the notation, let , , , is the identity matrix; First, the following nominal control law is designed for the nominal system: For the initial constrained following error of the system, the feedback control law is designed: For the system uncertainty part, the following reasonable assumptions are introduced: Assumption 1: For , is full rank; therefore It is reversible; Hypothesis 2: According to Hypothesis 1, for a given ,make There is a constant So that for all ,have in, represents the minimum eigenvalue of the matrix in brackets; Assumption 3: For all , there is an unknown constant vector and a known function , so there is The specific parameter perturbations are unknown, but constrains the overall boundaries of these uncertainties; A specific value of , an adaptive law is designed to update the estimated value in, , Is a vector No. Quantity; is a scalar constant; Based on the above assumptions about the uncertainty boundary, the following control components are designed to compensate for the uncertainty: in The total control input is thus defined as The controller consists of three parts: For nominal system control, Handling system initial constraint following deviation, With the help of Dealing with system uncertainty; In represents the control parameter, In is a constant matrix given in advance; the uncertainty parameter The value of is estimated by the adaptive law.

[0016] Preferably, in step S4, the Lyapunov function is selected as follows: Taking the derivative of the above formula with respect to time, we get the first-order derivative of the Lyapunov function as follows: By scaling the inequality, we can get: in, , make , Then, according to Lyapunov's min-max principle, we get the constraint following error and uncertainty parameter estimation error The uniform boundedness of is as follows: in is a given constant. And eventually uniformly bounded: . Uniform boundedness guarantees and Bounded, eventually uniform boundedness guarantees and It eventually converges to a neighborhood near 0 over time; according to The uniform boundedness of is bounded, and according to the properties of the state transformation, it is concluded that the correction error constraint is strictly guaranteed, thus ensuring the compact band performance of the original lateral error. The final uniform boundedness ensures that the tracking error and It will eventually converge to a neighborhood near 0.

[0017] A second aspect of the present invention provides a front-wheel steering and rear-wheel distributed drive vehicle path tracking system, comprising a planning layer, a control layer, a driving / braking force distribution, and a vehicle model; The planning layer is used to input the desired speed and desired path; The control layer receives the expected speed information and expected path information from the planning layer, and obtains the vehicle's lateral and longitudinal position, speed, yaw angle, steering angle, steering angular velocity and other information through the on-board inertial unit. The expected steering angle and active yaw torque are obtained through the proposed path tracking control algorithm and the distribution of driving and braking forces.

[0018] The driving / braking force distribution module distributes the active yaw torque to the desired driving force required by each driving wheel, and inputs it into the chassis motion control module in the vehicle model in combination with the desired steering angle calculated by the control layer to drive the vehicle movement.

[0019] Compared with the prior art, the present invention has the following beneficial technical effects: 1. The present invention is based on a vehicle dynamics model, adds an inertia time lag link to the vehicle dynamics model, and simultaneously considers uncertainty factors such as modeling errors and external interference, which is more in line with the changing characteristics of actual vehicle dynamics.

[0020] 2. In the lateral tracking error modeling, an auxiliary variable based on the conversion function is introduced to correct the original lateral error. The corrected lateral error is taken as the control object. The corrected lateral error is kept at a small value (the initial value is 0), so it can adapt to the large initial tracking error and avoid actuator saturation.

[0021] 3. The present invention obtains the error state without constraints by applying additional constraints and state transformation technology to the correction error. By ensuring the consistent bounded performance of the unconstrained system, the lateral tracking error can be strictly limited to a preset compact belt area at the transient and steady-state levels, ensuring the safety of the vehicle when performing path tracking tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic diagram of a front-wheel steering and rear-wheel distributed drive vehicle according to an embodiment of the present invention; Figure 2 A schematic diagram of a path tracking error according to an embodiment of the present invention; Figure 3 is a flow chart of a control method according to an embodiment of the present invention; Figure 4 A system diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0023] Example 1 The front-wheel steering and rear-wheel distributed drive vehicle dynamics model considered in the present invention is as follows: Figure 1 As shown, the vehicle longitudinal axis direction is Axis, the vehicle's horizontal axis direction is Axis, composition Body coordinate system, is the center of gravity of the vehicle. The center of gravity of the front wheelbase of the vehicle is , rear wheelbase center of gravity The front axle of the vehicle has a steering function. Assume that the steering angles of the two wheels on the front axle are consistent and set the front wheel steering angle Control interface. The two rear axle wheels are driving wheels, providing driving force / braking force , interface. is the active yaw moment, which is the sum of the torques generated by the driving force / braking force generated by the driving wheels on the left and right sides of the rear axle around the center of mass. After that, it is allocated to the left and right driving wheels of the rear axle according to the corresponding allocation rules. In addition, the autonomous driving vehicle is equipped with a positioning module, a perception module, a decision-making planning module and a control module. It is also equipped with a steering angle sensor and a wheel speed sensor to obtain the actual steering angle of the vehicle. and steering angular velocity The planning module provides the control module with the desired path information and the curvature information of the desired path. The control module obtains the active yaw torque and the desired steering angle control value according to the designed path tracking algorithm. The active yaw torque is input to the driving wheel and steering mechanism in the chassis module together with the steering angle after the driving force distribution. The path tracking error model is as follows: Figure 2 As shown, The lateral tracking error is the lateral distance from the center of gravity of the vehicle to the nearest desired path point. The heading error is the difference between the vehicle heading angle and the direction of the nearest desired path point. The vehicle can obtain the relative position between the vehicle and the desired path through calculation. In addition, the vehicle can obtain its own motion information through the positioning module sensor: longitudinal speed , lateral speed , yaw angle , yaw rate .

[0024] The design steps of the vehicle lateral path tracking controller proposed by the present invention that meets the preset performance are as follows: Constructing a vehicle lateral dynamics model The lateral dynamics of an actual front-wheel steering and rear-wheel distributed drive vehicle is affected by many factors. In order to balance the simplicity and accuracy of the model analysis, the following reasonable assumptions are made: (1) Assume that the frame and suspension are rigid structures and ignore their weak nonlinear effects. In addition, the vehicle roll and pitch dynamics can be ignored, and the focus is on the vehicle's lateral and yaw dynamics.

[0025] (2) It is assumed that the side plate stiffness and sideslip angle of the coaxial left and right tires on the front and rear axles are the same, and the steering angles of the left and right steering wheels on the front axle are the same.

[0026] Since any mechanical system has time lag, especially the vehicle steering mechanism, after the controller sends a steering signal, the steering mechanism needs a certain amount of time to complete the wheel steering operation. The delay time here is called the steering mechanism inertia time lag. The existence of the time lag problem poses a great challenge to the safety of autonomous vehicles. This factor cannot be ignored when designing the controller. Therefore, combining Newtonian mechanics and the rigid body rotation theorem, the vehicle lateral dynamics model with a second-order inertia time lag link is established as follows:

[0027] (1) (2) (3) in, is the vehicle mass, is the vehicle moment of inertia, and is a time-varying bounded uncertainty parameter caused by model errors and external disturbances. is the desired front wheel steering angle of the vehicle. and It is the system parameter of the time-delay second-order steering system. Its specific value needs to be obtained by steering test calibration in advance. Other dynamic parameters ~ The specific expression is as follows:

[0028] (4) (5) in, is the front wheel cornering stiffness, is the rear wheel cornering stiffness. In practice, the dynamic parameters in (4) and (5) (such as ) are difficult to identify because they may be time-varying. Therefore, we assume that the parameters ~ All by nominal part and the uncertain part which is unknown and changes with time composition.

[0029] Constructing a path-following kinematic model The vehicle obtains its own longitudinal and transverse coordinates and yaw angle and other posture information in the geodetic coordinate system through the inertial measurement unit (IMU). The vehicle obtains the longitudinal and transverse coordinates, yaw angle and road curvature of the desired path point through the upper planning module. The yaw angle error of the vehicle relative to the desired path can be expressed as:

[0030] (6) in, is the ideal yaw angle at the desired path point. According to the road curvature, the ideal yaw angle change rate can be expressed as: (7) Then the first-order and second-order forms of e_φ are obtained as follows: (8) (9) Assuming that the rate of change of the road curvature is small, the The term can be ignored. The lateral error is defined as the shortest distance between the vehicle's current position and the nearest desired path point. According to the kinematic relationship, the first-order and second-order forms of the lateral error are as follows:

[0031] (11) (12) To facilitate the subsequent controller design, the vehicle lateral dynamics (1)-(6) and equations (9)-(12) are integrated to establish the state space equation based on error dynamics as follows: (13) Among them, the vector is the state variable, For control input. Other parts The specific expression is as follows:

[0032] (14) Note that the system with steering mechanism time lag is a typical underactuated system. The system has three state variables (lateral error, heading error, actual steering angle), but only two control inputs. This underactuated characteristic combined with vehicle dynamics uncertainty makes vehicle lateral path tracking control a challenge. The robust control method based on constraint following theory to be designed next can handle the above problems.

[0033] 2.2.3 Constructing Path Tracking Task Constraints (1) Introducing auxiliary variables The actual road has boundaries, so the lateral offset of the vehicle's path tracking must be within the road safety range. We express this safety constraint as the following inequality: (15) in, and are the minimum and maximum allowed lateral offsets, respectively.

[0034] First, introduce Definition 1: If the following conditions are met, it is a conversion function: (1) And in It is monotonically increasing on Keep on ,in is the preset time.

[0035] (2) for And for all All have boundaries, is the order of the controlled system.

[0036] The initial position of the vehicle may be far from the nearest point on the desired path (i.e. Too large) will cause the controller to produce saturated control force. Long-term actuator saturation will deteriorate the control performance and even cause actuator failure, leading to safety accidents. Therefore, the introduction of auxiliary variables To correct the lateral error, the corrected error is as follows:

[0037] (16) in, , conversion function According to Definition 1, we select: (17) Based on the above corrections, combined with (15) and (16), we can obtain that the corrected lateral error has the following properties: (18) Combined with formula (15), in order to ensure that the original lateral error still satisfies the bilateral inequality after correction, the following constraints are imposed on the corrected lateral error: (19) Substituting equation (16) into equation (19) and combining it with equation (15), in order to strictly limit the original lateral error within the preset strip area, equation (19) needs to be restated as: (20) Among them, the preset performance boundary parameters and The following inequality is satisfied: (twenty one) This aims to limit the original lateral error to a more stringent band area, which also means that the safety of the vehicle is strictly guaranteed during the transient adjustment of the lateral error. Later we will combine the state transformation technology to achieve this goal.

[0038] (2) State Transition This section proposes a transformation method based on the barrier transformation function to transform the bounded state Convert to unbounded state , the barrier transformation function based on the tangent function aims to achieve the following transformation: (twenty two) (twenty three) in, By taking the derivative of equation (23) with respect to time t, we obtain: (twenty four) (25) In addition, we use To represent the vehicle yaw angle error, that is ,make , so the controlled state quantity Can be converted to a new state Then we can get the following formula:

[0039] (26) in (27) Since for all , both have Established, therefore is full rank. After the state transformation, about the unconstrained variables Status The kinetic equation can be expressed as:

[0040] (28) in (29) because and are all full rank, so is also full rank, that is is also reversible. In addition, the premise of the above transformation is The initial value of satisfies the constraint condition of formula (19). Obviously, after introducing the auxiliary variable correction, this condition is naturally satisfied, because Always holds true. and are all monotonically increasing functions. Combined with Under the condition of yes Through this transformation, The bilateral inequality constraint of is equivalent to the new state Next, we aim to design a controller that ensures the transformation state in the presence of time-varying uncertainty in vehicle dynamics. The uniform boundedness and eventual uniform boundedness of .

[0041] (3) Establishing equality constraints The goal of the path tracking task for autonomous vehicles is to make the path tracking error converge to 0, that is, the vehicle drives on the desired path point. This convergence performance can be expressed by the following first-order and second-order equality constraints on the system state:

[0042] (30) (31) Solving the differential equations (30) and (31), the corrected lateral error and yaw angle error will be and , and approaches 0 over time. Based on the characteristics of state transformation, the same constraints are imposed on the transformation state:

[0043] (32) (33) Write the above equality constraint in matrix form , and order , which represents the constraint following error.

[0044] in, , .

[0045] The second-order form of the constraints is: in, .

[0046] Adaptive robust controller design This section will introduce the steps involved in designing a robust controller based on the constraint following theory. First, the servo control law is designed for the nominal system without considering the initial constraint following error and uncertainty of the system. Second, the feedback part control law and the uncertainty compensation part control law are designed to compensate for the initial deviation and uncertainty of the system.

[0047] The parameter matrix in equation (28) is decomposed into the following nominal part and uncertain part: (34) The matrix with horizontal lines is the known nominal part, and the other matrices represent the unknown uncertain parts. To simplify the notation, let , , , is the identity matrix.

[0048] First, the following nominal control law is designed for the nominal system: (35) For the initial constrained following error of the system, the feedback control law is designed: (36) For the system uncertainty part, the following reasonable assumptions are introduced: Assumption 1: For , is full rank. So It is reversible.

[0049] Hypothesis 2: According to Hypothesis 1, for a given ,make (37) There exists a constant (possibly unknown) So that for all ,have (38) in, Represents the smallest eigenvalue of the matrix in brackets.

[0050] Assumption 3: For all , there is an unknown constant vector and a known function , so there is (39) The specific parameter perturbations are unknown, but This limits the overall bounds of these uncertainties. A specific value of , an adaptive law is designed to update the estimated value

[0051] (40) in, ( Is a vector No. Quantity); is a scalar constant.

[0052] Based on the above assumptions about uncertainty boundaries, the following control components are designed to compensate for the uncertainty: (41) in (42) (43) The total control input is thus defined as (44) The controller consists of three parts: For nominal system control, Handling system initial constraint following deviation, With the help of Dealing with system uncertainty. In represents the control parameter, In is a constant matrix given in advance. Uncertainty parameters The value of is estimated by the adaptive law.

[0053] Stability and safety analysis The Lyapunov stability theory is used to analyze the controller stability and the performance of the constrained following error, thereby analyzing the performance of the lateral tracking error. The Lyapunov function is selected as follows:

[0054] (45) Taking the derivative of the above formula with respect to time, we get the first-order derivative of the Lyapunov function as follows: (46) Combining equations (28) and (35)-(41), and scaling the inequality, we can obtain: (47) in, , make , Then, according to Lyapunov's min-max principle, we get the constraint following error and uncertainty parameter estimation error The uniform boundedness and eventual uniform boundedness of . Uniform boundedness guarantees and Bounded, eventually uniform boundedness guarantees and Over time, it eventually converges to a neighborhood near 0. The uniform boundedness of is bounded, and according to the properties of the state transformation, we can conclude that constraint (19) is strictly guaranteed, thus ensuring the original lateral error The compact band performance of . The eventual uniform boundedness ensures that the tracking error and It will eventually converge to a neighborhood near 0.

[0055] Driving / braking force distribution The control quantity obtained by the control law (44) is In It is composed of the driving force of the left and right rear drive wheels and cannot be directly input into the chassis for motion control. It needs to be distributed according to the following driving / braking force rules. Assigned to left and right rear drive wheels: (48) in, The total driving force is calculated for the longitudinal speed control of the vehicle, and the longitudinal control is not taken into account in the present invention. The wheelbase of the vehicle.

[0056] The present invention is different from the existing lateral path tracking control algorithm in terms of vehicle dynamics modeling and path tracking task modeling. Inertial time lag links and dynamic uncertainty factors are integrated into the dynamic model, which is more in line with the actual vehicle motion characteristics. The introduction of auxiliary variables based on the conversion function corrects the original lateral error so that its initial value is 0, which can adapt to the working conditions of large initial tracking errors and avoid actuator saturation. After state transformation, by realizing the consistent boundedness and ultimate consistent boundedness of the unconstrained system, the preset compact belt area performance of the tracking error is strictly guaranteed, which improves the safety of the autonomous driving vehicle when performing path tracking tasks.

[0057] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle, characterized in that: The specific steps include: S1. Establish a nonlinear vehicle dynamics model with front-wheel steering and rear-wheel distributed drive; S11. Constructing a vehicle lateral dynamics model; S12, constructing a path tracking operation model; S2, establish path tracking task constraints; S21, introduce auxiliary variables to correct the bilateral constraints of lateral errors; S22, perform state bijective transformation to convert the bounded state into an unbounded state; S23, establishing path tracking equality constraints; S3. Design an adaptive robust controller based on task constraints; S31. Design a servo control law for the nominal system; S32, design the error feedback partial control law; S33. Design an adaptive compensation partial control law to compensate for the initial deviation and uncertainty of the system; S4. Stability and constrained following error performance of adaptive robust controller based on Lyapunov analysis; S5. Use the controller in step S3 to control the vehicle to perform path tracking.

2. The path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle according to claim 1, characterized in that: The nonlinear vehicle dynamics model of front-wheel steering and rear-wheel distributed drive established in step S1 includes: The vehicle longitudinal axis is Axis, the vehicle's horizontal axis direction is Axis, composition Body coordinate system, is the center of gravity of the vehicle; the center of gravity of the front wheelbase of the vehicle is , rear wheelbase center of gravity ; The front axle of the vehicle has a steering function. Assume that the steering angles of the two wheels on the front axle are consistent and set the front wheel steering angle Control interface; the two rear axle wheels are driving wheels, providing driving force / braking force , interface; The model also includes a positioning module, a perception module, a decision-making planning module and a control module; it is also equipped with a steering angle sensor and a wheel speed sensor to obtain the actual steering angle of the vehicle. and steering angular velocity ; The planning module provides the control module with the desired path information and the curvature information of the desired path The control module obtains the active yaw torque and the desired steering angle control amount according to the designed path tracking algorithm. The active yaw torque is input into the driving wheel and steering mechanism in the chassis module together with the steering angle after the driving force distribution.

3. The path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle according to claim 2, characterized in that: In step S11, a vehicle lateral dynamics model with a second-order inertia time lag is established as follows: in, is the vehicle mass, is the vehicle moment of inertia, and is a time-varying bounded uncertainty parameter caused by model errors and external disturbances; is the desired front wheel steering angle of the vehicle; and It is the system parameter of the time-delay second-order steering system, and its specific value needs to be obtained through steering test calibration; Other kinetic parameters The specific expression is as follows: in, is the front wheel cornering stiffness, is the rear wheel cornering stiffness; in practice, the dynamic parameters in the above formula are It is difficult to identify, so we assume that the parameter All by nominal part and the uncertain part which is unknown and changes with time composition.

4. The path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle according to claim 1, characterized in that: In step S12, the vehicle obtains its own longitudinal and transverse coordinates and yaw angle and other posture information in the geodetic coordinate system through the inertial measurement unit, and the vehicle obtains the longitudinal and transverse coordinates, yaw angle and road curvature information of the desired path point through the upper planning module; the yaw angle error of the vehicle relative to the desired path can be expressed as: in, is the heading angle error, is the yaw angle; is the ideal yaw angle at the desired path point. According to the road curvature, the ideal yaw angle change rate can be expressed as: is the longitudinal velocity; then we get The first-order and second-order forms of are as follows: The lateral error is defined as the shortest distance between the vehicle's current position and the nearest desired path point. According to the kinematic relationship, the first-order and second-order forms of the lateral error are as follows: The state space equation based on error dynamics is established as follows: Among them, the vector is the state variable, is the control input; other coefficient matrices The specific expression is as follows: 。 5. The path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle according to claim 1, characterized in that: In step S21, the path tracking lateral offset of the vehicle must meet the following boundary conditions: in, and are the minimum and maximum lateral offsets allowed, respectively; Introduce definition A: If the following conditions are met, it is a conversion function: A. And in It is monotonically increasing on Keep on ,in is the preset time; B. for And for all All have boundaries, is the order of the controlled system; The initial position of the vehicle may be far from the nearest point of the desired path, i.e. Too large; introduce auxiliary variables To correct the lateral error, the corrected error is as follows: in, , conversion function According to the definition A is as follows: Based on the above correction, the corrected lateral error has the following properties: The following constraints are imposed to correct the lateral error: The above formula can be restated as: Among them, the preset performance boundary parameters and The following inequality is satisfied: The original lateral error is restricted to a tighter band.

6. The path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle according to claim 1, characterized in that: In step S22, the bounded state Convert to unbounded state , the barrier transformation function based on the tangent function aims to achieve the following transformation: in, Derivative the above formula with respect to time t, we get: use To represent the vehicle yaw angle error, that is ,make , so the controlled state quantity Can be converted to a new state ; The following formula is obtained: in Since for all , both Established, therefore is full rank; after the state transformation, about the unconstrained variables Status The kinetic equation can be expressed as: in because and are all full rank, so is also full rank, that is It is also reversible.

7. The path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle according to claim 1, characterized in that: In step S23, the goal of the path tracking task of the autonomous driving vehicle is to make the path tracking error converge to 0, that is, the vehicle travels on the desired path point; the convergence performance is converted into the following first-order and second-order equality constraints on the system state: Solving the differential equations above, the corrected lateral error and yaw angle error will be and , and approaches 0 over time; based on the characteristics of state transformation, the same constraints are imposed on the transformation state: Write the above equality constraint in matrix form , and order , which represents the constraint following error; in, , ; The second-order form of the constraints is: in, .

8. The path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle according to claim 1, characterized in that: In step S3, after the state transformation, the unconstrained variables The parameter matrix of the dynamic equation of state Z is decomposed into the following nominal part and uncertain part: The matrix with horizontal lines is the known nominal part, and the other matrices represent the unknown uncertain parts. To simplify the notation, let , , , is the identity matrix; First, the following nominal control law is designed for the nominal system: For the initial constrained following error of the system, the feedback control law is designed: For the system uncertainty part, the following reasonable assumptions are introduced: Assumption 1: For , is full rank; therefore It is reversible; Hypothesis 2: According to Hypothesis 1, for a given ,make There is a constant So that for all ,have in, represents the minimum eigenvalue of the matrix in brackets; Assumption 3: For all , there is an unknown constant vector and a known function , so there is The specific parameter perturbations are unknown, but constrains the overall boundaries of these uncertainties; A specific value of , an adaptive law is designed to update the estimated value in, , Is a vector No. Quantity; is a scalar constant; Based on the above assumptions about uncertainty boundaries, the following control components are designed to compensate for the uncertainty: in The total control input is thus defined as The controller consists of three parts: For nominal system control, Handling system initial constraint following deviation, With the help of Dealing with system uncertainty; In represents the control parameter, In is a constant matrix given in advance; the uncertainty parameter The value of is estimated by the adaptive law.

9. The path tracking method for a front-wheel steering and rear-wheel distributed drive vehicle according to claim 1, characterized in that: In step S4, the Lyapunov function is selected as follows: Taking the derivative of the above formula with respect to time, we get the first-order derivative of the Lyapunov function as follows: By scaling the inequality, we can get: in, , make , Then, according to Lyapunov's min-max principle, we get the constraint following error and uncertainty parameter estimation error The uniform boundedness and eventual uniform boundedness of and Bounded, eventually uniform boundedness guarantees and It eventually converges to a neighborhood near 0 over time; according to The uniform boundedness of is bounded, and the eventual uniform boundedness guarantees the tracking error and It will eventually converge to a neighborhood near 0.

10. A front-wheel steering and rear-wheel distributed drive vehicle path tracking system, characterized in that: Includes planning layer, control layer, driving / braking force distribution and vehicle model; The planning layer is used to input the desired speed and desired path; The control layer receives the expected speed information and expected path information from the planning layer, and obtains the vehicle's lateral and longitudinal position, speed, yaw angle, steering angle, steering angular velocity and other information through the vehicle-mounted inertial unit. The expected steering angle and active yaw moment are obtained through the proposed path tracking control algorithm and the distribution of driving and braking forces. The driving / braking force distribution module distributes the active yaw torque to the desired driving force required by each driving wheel, and inputs it into the chassis motion control module in the vehicle model in combination with the desired steering angle calculated by the control layer to drive the vehicle movement.

Images