A reconfigurable distributed drive multi-axis vehicle trajectory tracking method and carrier device

Abstract

The present invention discloses a reconfigurable distributed drive multi-axis vehicle trajectory tracking method and carrier. The method includes real-time acquisition of the longitudinal and lateral motion speeds and yaw angular velocity of the vehicle body; obtaining the vehicle's position information and target path information, and matching the trajectory tracking target point; establishing a three-degree-of-freedom vehicle dynamics model for trajectory tracking; using a sliding mode control method to solve the longitudinal force required for the vehicle based on the vehicle's longitudinal dynamics model to achieve the effect of tracking a specified speed; obtaining a tracking error model, and calculating the wheel angle and vehicle yaw moment based on a nonlinear model predictive control algorithm; allocating the longitudinal tire force to each wheel based on the tire adhesion definition and the vehicle longitudinal force and yaw moment requirements output by the upper-level controller, and converting it into motor torque; and controlling the vehicle's steering system and wheel motors to execute wheel steering angle and torque commands. The present invention is beneficial for improving the vehicle's handling stability and trajectory tracking accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of vehicle dynamics control, and in particular relates to a reconfigurable distributed drive multi-axis vehicle trajectory tracking method and a carrying device. Background Art

[0002] Autonomous driving is one of the hottest technologies today, offering unique advantages in improving road safety, enhancing travel convenience, and promoting the new energy economy. In particular, for heavy-duty commercial transport vehicles operating fixed routes and specialized missions, the integration of autonomous driving technology will significantly enhance traffic efficiency, optimize road space utilization, reduce transportation costs, and help alleviate urban traffic pressures.

[0003] The core algorithmic system for autonomous driving encompasses three key dimensions: environmental perception, decision-making and planning, and control execution. Trajectory tracking control, as a downstream component of the autonomous driving algorithm, has a decisive impact on the overall system. Accurate and secure trajectory tracking is crucial for ensuring that autonomous vehicles can smoothly follow their pre-set paths, avoid obstacles, and safely reach their destinations.

[0004] The accuracy of existing autonomous driving in trajectory tracking and speed tracking needs to be further improved. Summary of the Invention

[0005] Purpose of the invention: The present invention provides a reconfigurable distributed drive multi-axis vehicle trajectory tracking method and carrier, which can be integrated into the autonomous driving technology of commercial vehicles to improve the trajectory tracking speed and tracking effect.

[0006] Reconfigurable distributed-drive multi-axle vehicles offer inherent advantages as actuators for trajectory tracking. Distributed-drive multi-axle electric vehicles with all-wheel steering replace the mechanical connections of traditional centralized-drive vehicles with a drive-by-wire system. This allows for independent control and rapid response of steering and torque at each wheel, providing a more convenient platform for vehicle trajectory tracking control. Furthermore, due to the variable number of axles, reconfigurable distributed-drive multi-axle vehicles can adjust the total number of axles to suit load requirements. This feature increases the vehicle's carrying capacity and further reduces freight costs.

[0007] Technical solution: The present invention provides a reconfigurable distributed drive multi-axis vehicle trajectory tracking method and a carrier device, comprising the following steps:

[0008] (1) Obtain the longitudinal and lateral motion speeds and yaw angular velocity of the vehicle body in real time;

[0009] (2) Obtain the vehicle's location information, target path information, and match the appropriate trajectory to track the target point;

[0010] (3) Establish a three-degree-of-freedom dynamic model of the vehicle for trajectory tracking, where the three degrees of freedom include the longitudinal, lateral, and yaw directions of the vehicle;

[0011] (4) Based on the longitudinal dynamics model of the vehicle, the sliding mode control method is used to solve the longitudinal force required by the vehicle to achieve the effect of tracking the specified speed;

[0012] (5) Based on the Frenet coordinate system and the vehicle dynamics model, the tracking error model is obtained, and the wheel angle and vehicle yaw moment are calculated based on the nonlinear model predictive control algorithm;

[0013] (6) According to the definition of tire adhesion and the requirements of the longitudinal force and yaw moment of the vehicle output by the upper controller, the quadratic programming method is used to optimally distribute the longitudinal tire force of each wheel and convert it into motor torque.

[0014] (7) Control the vehicle's steering system and wheel motors to execute wheel steering angle and torque commands.

[0015] Furthermore, the step (3) includes:

[0016] like Figure 2 As shown in Figure 2, the lateral dynamics and yaw dynamics models of the vehicle are as follows:

[0017]

[0018] The lateral force of the wheel is expressed by the following formula:

[0019]

[0020] T is the coefficient matrix. When the axis i exists, t i is a positive integer 1, otherwise it is 0. The expression of α is as follows:

[0021]

[0022] The symbols and numbers in the above formula are explained: j = 1 and j = 2 represent the left and right wheels respectively, i represents the i-th axle counting from the front end of the vehicle, m is the total mass of the vehicle, v y is the lateral velocity at the center of mass of the vehicle in the body coordinate system, v x is the longitudinal velocity at the center of mass of the vehicle in the body coordinate system, ω r is the vehicle's yaw rate, I is the vehicle's moment of inertia around the z axis, l i represents the distance from the i-th axis to the center of mass of the vehicle, F yij Indicates the lateral force provided by a single tire, F yi represents the lateral force provided by a single axle, α is the side slip angle of the wheel, δ is the wheel turning angle, d is the wheelbase, ΔM zis the additional direct yaw moment of the vehicle, k is the tire cornering stiffness, represents the differential of yaw rate with respect to time t;

[0023] The longitudinal dynamics model of the vehicle is as follows:

[0024]

[0025] Among them, F x is the sum of the longitudinal forces acting on the vehicle, C D Indicates the drag coefficient, A f is the frontal area of ​​the vehicle, ρ0 is the air density, and d1 is the interference term.

[0026] Furthermore, the step (4) includes:

[0027] The velocity tracking error is defined as:

[0028] e v =v x -v xd

[0029] Introduce the integral term into the sliding mode function:

[0030]

[0031] The Lyapunov function is defined as:

[0032]

[0033] In order to satisfy the Lyapunov stability condition, that is, The control rate of the designed vehicle longitudinal force is:

[0034]

[0035] Explain the symbols and numbers in the above formula: c, k2, ε, D are all positive constants, v xd is the target longitudinal velocity, v x is the actual longitudinal velocity, e v is the difference between the actual longitudinal velocity and the desired velocity, s is the sliding mode surface function, and in order to alleviate the chattering phenomenon that is easily produced by ordinary sliding mode controllers, the hyperbolic tangent function tanh is used instead of the ordinary switching function:

[0036]

[0037] Furthermore, the vehicle trajectory tracking using NMPC in step (5) includes:

[0038] The vehicle tracking error model is obtained through the Frenet coordinate system:

[0039]

[0040] Taking the derivatives of both sides of the above equations, we get:

[0041]

[0042] Combined with the vehicle's lateral dynamics model:

[0043]

[0044] Take the state variables of the system as:

[0045]

[0046] The control quantity is:

[0047] u=[δ 11 δ 12 δ 21 δ 22 δ 31 δ 32 δ 41 δ 42 δ 51 δ 52 δ 61 δ 62 ΔM z ] T

[0048] According to the above derivation, the state space equation of the vehicle power system can be obtained:

[0049]

[0050] The cost function for the path tracing stage is set to:

[0051]

[0052] Then at time t, the optimal control problem of NMPC can be expressed as follows:

[0053]

[0054] u min ≤u(t+i)≤u max

[0055] Explain the symbols and numbers in the above formula: N is the prediction time domain, J represents the cost function, and the weight matrix of the cost function is Q, R1, R2, is the vehicle's yaw angle, e d Represents the error between the vehicle's center of mass and the target point, is the difference between the vehicle's yaw angle and the desired yaw angle, is the desired yaw angle, is the rate of change of the heading angle of the target point, k1 is the curvature of the path point, is the rate of change of the horizontal coordinate at the path point in the Frenet coordinate system, Δu represents the rate of change of the control quantity, u min and u max are the maximum and minimum values ​​of the control quantity, respectively. l(x,u), g(x,u), and f(x,u) represent functions of x and u.

[0056] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. The present invention takes into account the variable number of axles in a reconfigurable multi-axle distributed drive vehicle and adopts a coefficient matrix method to enable the vehicle dynamics model to adapt to changes in the number of axles.

[0058] 2. The present invention proposes an integrated trajectory tracking control strategy based on NMPC and sliding mode variable structure, which can improve trajectory tracking accuracy while also ensuring driving stability to a certain extent; 3. The present invention can be integrated into the automatic driving system of commercial vehicles to improve tracking effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flow chart of an embodiment of the present invention;

[0060] Figure 2 It is an application object diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The present invention will be further described in detail below with reference to the accompanying drawings.

[0062] The reconfigurable multi-axle distributed drive vehicle has an adjustable number of axles, and each wheel has both steering and drive functions. It also has radar, GPS, and other features, enabling autonomous driving.

[0063] During the autonomous driving process of a reconfigurable multi-axle distributed drive vehicle, the decision-making and planning layer provides the specified target trajectory information. At the lower level, the controller, replacing the human operator, completes a series of driving operations to ensure that the vehicle follows the specified trajectory and maintains the desired speed as closely as possible. Leveraging the actuator redundancy of multi-axle commercial vehicles for coordinated multi-actuator control improves trajectory tracking. To adapt to the variable number of axles in reconfigurable multi-axle distributed drive vehicles and ensure effective trajectory and speed tracking, it is necessary to establish a precise dynamic model and design a corresponding control strategy. This distributes the steering angle and torque information obtained by the controller to each wheel until the autonomous driving process is complete.

[0064] Figure 2A six-axle commercial vehicle is presented as an example of the reconfigurable multi-axle distributed drive vehicle of the present invention. The invention first uses an onboard IMU to obtain yaw rate information and a state estimation module to obtain the vehicle's longitudinal and lateral velocity information. The onboard GPS module then acquires the vehicle's position in real time, using this information to locate the target point for trajectory tracking within the upper-level planned path. A reconfigurable longitudinal and lateral dynamics model of the vehicle is then constructed based on the vehicle body state information. A tracking error model is then designed to transform the trajectory tracking problem into an NMPC optimization problem, and the velocity error model is converted into the control rate of a sliding mode controller to solve for the steering angle of each wheel and the yaw moment of the vehicle. Finally, a tire force distribution strategy is designed to obtain torque commands for each wheel motor. The steering angle and torque information are then distributed to the corresponding actuators for execution until the autonomous driving process ends.

[0065] With the help of Figure 2 The scenario describes the implementation process of the present invention in detail. Figure 1 As shown, the energy-optimized braking speed optimization method for an intelligent connected electric vehicle according to the present invention specifically includes the following steps:

[0066] Step 1: Obtain the vehicle body motion state; use the vehicle-mounted sensors to obtain the vehicle body's longitudinal and lateral motion speeds and yaw angular velocity in real time;

[0067] The information acquisition method is:

[0068] The longitudinal vehicle speed information is obtained by the on-board wheel speed sensor, which is installed on the vehicle and is an essential sensor for the vehicle; the yaw angular velocity is obtained by the on-board inertial measurement unit; and the vehicle lateral speed is obtained by the vehicle state estimation module.

[0069] Step 2: Target path point matching: Use the vehicle's GPS to obtain the vehicle's location information, extract the target path information from the trajectory planning layer, and match the appropriate trajectory tracking target point.

[0070] The information acquisition method is:

[0071] The vehicle's location information is obtained by the on-board GPS module; the target path information is obtained through the trajectory planning layer of the autonomous driving program.

[0072] The third step is adaptive dynamic modeling; establish a three-degree-of-freedom dynamic model of the vehicle for trajectory tracking, which includes the longitudinal, lateral and yaw directions of the vehicle and can adapt to changes in the number of vehicle axles.

[0073] The dynamic model of a multi-axle vehicle is:

[0074] Define j = 1 and j = 2 to represent the left and right wheels of the vehicle respectively, i represents the i-th axle counting from the front end of the vehicle, the total mass of the vehicle is m, and the lateral velocity at the center of mass of the vehicle in the body coordinate system is vy , the longitudinal velocity at the center of mass of the vehicle in the body coordinate system is v x , the vehicle's center of mass sideslip angle is β, and the vehicle's yaw rate is ω r , the moment of inertia of the vehicle around the z-axis is I,l i represents the distance from the i-th axle to the center of mass of the vehicle, and the lateral force F provided by a single tire yij , the lateral force provided by a single shaft is F yi , the wheel slip angle is α, the wheel steering angle is δ, the wheelbase is d, and the additional direct yaw moment of the vehicle is ΔM z The tire cornering stiffness is k, and the differential of the yaw rate with respect to time t is

[0075] The lateral dynamics and yaw dynamics models of the vehicle are as follows:

[0076]

[0077] The lateral force of the wheel is expressed by the following formula:

[0078]

[0079] T is the coefficient matrix. When the axis i exists, t i is a positive integer 1, otherwise it is 0. The expression of α is as follows:

[0080]

[0081] The total longitudinal force on the vehicle is defined as F x , the drag coefficient is C D , the frontal area of ​​the vehicle is A f , the air density is ρ0, and the interference term is d1.

[0082] The longitudinal dynamics model of the vehicle is as follows:

[0083]

[0084] The fourth step is to design a speed tracking controller. Based on the vehicle's longitudinal dynamics model, a sliding mode control method is used to solve the required longitudinal force of the vehicle to achieve the effect of tracking the specified speed.

[0085] Define c, k2, ε, and D as positive constants, and the target longitudinal velocity is v xd , the actual longitudinal velocity is v x , the difference between the actual longitudinal velocity and the expected velocity is e v , the sliding surface function is s, and the symbol of the hyperbolic tangent function is tanh.

[0086] The velocity tracking error is defined as:

[0087] e v =v x -v xd

[0088] Introduce the integral term into the sliding mode function:

[0089]

[0090] The Lyapunov function is defined as:

[0091]

[0092] In order to satisfy the Lyapunov stability condition, that is, The control rate of the longitudinal force of the designed vehicle is

[0093]

[0094] The fifth step is to design the trajectory tracking controller; based on the Frenet coordinate system and the vehicle dynamics model, the tracking error model is obtained, and the wheel angle and the yaw moment of the whole vehicle are calculated based on the nonlinear model predictive control algorithm.

[0095] N is the prediction time domain, J represents the cost function, and the weight matrix of the cost function is Q, R1, R2, is the vehicle's yaw angle, e d Represents the error between the vehicle's center of mass and the target point, is the difference between the vehicle's yaw angle and the desired yaw angle, is the desired yaw angle, is the rate of change of the heading angle of the target point, k1 is the curvature of the path point, is the rate of change of the horizontal coordinate at the path point in the Frenet coordinate system, Δu represents the rate of change of the control quantity, u min and u max are the maximum and minimum values ​​of the control variables, respectively; l(x,u), g(x,u), and f(x,u) represent functions of x and u.

[0096] The vehicle tracking error model is obtained through the Frenet coordinate system:

[0097]

[0098] Taking the derivatives of both sides of the above equations, we get:

[0099]

[0100] Combined with the vehicle's lateral dynamics model:

[0101]

[0102] Take the state variables of the system as:

[0103]

[0104] The control quantity is:

[0105] u=[δ 11 δ 12 δ 21 δ 22 δ 31 δ 32 δ 41 δ 42 δ 51 δ 52 δ 61 δ 62 ΔM z ] T

[0106] According to the above derivation, the state space equation of the vehicle power system can be obtained:

[0107]

[0108] Set the path tracing stage cost function to

[0109]

[0110] Then at time t, the optimal control problem of NMPC can be expressed as follows:

[0111]

[0112] u min ≤u(t+i)≤u max

[0113] The sixth step is to optimize tire force distribution. Based on the definition of tire adhesion and the vehicle longitudinal force and yaw moment requirements output by the upper-level controller, a quadratic programming method is used to optimally distribute the longitudinal tire force on each wheel and convert it into motor torque.

[0114] End: The multi-axle electric vehicle operates according to the steering angle and torque given by the integrated controller until the autonomous driving mode is exited.

[0115] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless defined as such herein, will not be interpreted in an idealized or overly formal sense.

[0116] The meaning of "and / or" in this application means that both situations where each exists alone or both exist at the same time are included.

[0117] With the above-described preferred embodiments of the present invention as a guide, and with reference to the above description, relevant personnel are fully capable of making various changes and modifications without departing from the technical scope of this invention. The technical scope of this invention is not limited to the contents of the specification and must be determined according to the scope of the claims.

Claims

1. A reconfigurable distributed drive multi-axis vehicle trajectory tracking method, characterized in that: The following steps are involved: (1) Obtain the longitudinal and lateral motion speeds and yaw angular velocity of the vehicle body in real time; (2) Obtain the vehicle's location information, target path information, and match the trajectory to track the target point; (3) Establish a three-degree-of-freedom dynamic model of the vehicle for trajectory tracking, where the three degrees of freedom include the longitudinal, lateral, and yaw directions of the vehicle; (4) Based on the longitudinal dynamics model of the vehicle, the sliding mode control method is used to solve the longitudinal force required by the vehicle to achieve the effect of tracking the specified speed; (5) Based on the Frenet coordinate system and the vehicle dynamics model, the tracking error model is obtained, and the wheel angle and vehicle yaw moment are calculated based on the nonlinear model predictive control algorithm; (6) Based on the definition of tire adhesion and the vehicle longitudinal force and yaw moment requirements output by the upper controller, the quadratic programming method is used to optimally distribute the longitudinal tire force of each wheel and convert it into motor torque; (7) Control the vehicle's steering system and wheel motors to execute wheel steering angle and torque commands; The lateral dynamics and yaw dynamics models of the vehicle in step (3) are as follows: The symbols and numbers in the above formula are explained: j = 1 and j = 2 represent the left and right wheels respectively, i represents the i-th axle counting from the front end of the vehicle, m is the total mass of the vehicle, v y is the lateral velocity at the center of mass of the vehicle in the body coordinate system, v x is the longitudinal velocity at the center of mass of the vehicle in the body coordinate system, ω r is the vehicle's yaw rate, I is the vehicle's moment of inertia around the z axis, l i represents the distance from the i-th axis to the center of mass of the vehicle, F yij Indicates the lateral force provided by a single tire, F yi Indicates the lateral force provided by a single shaft, ΔM z is the additional direct yaw moment of the vehicle, represents the differential of yaw rate with respect to time t; The step (5) comprises: The vehicle tracking error model is obtained through the Frenet coordinate system: Taking the derivatives of both sides of the above equations, we get: Combined with the vehicle's lateral dynamics model: Take the state variables of the system as: The control quantity is: u=[δ 11 d 12 d 21 d 22 d 31 d 32 d 41 d 42 d 51 d 52 d 61 d 62 DM z ] T The state space equations of the vehicle dynamics system: The cost function for the path tracing stage is set to: Then at time t, the optimal control problem of nonlinear model predictive control NMPC is expressed as follows: u min ≤u(t+i)≤u max Explain the symbols and numbers in the above formula: N is the prediction time domain, J represents the cost function, and the weight matrix of the cost function is Q, R1, R2, is the vehicle's yaw angle, e d Represents the error between the vehicle's center of mass and the target point, is the difference between the vehicle's yaw angle and the desired yaw angle, is the desired yaw angle, is the rate of change of the heading angle of the target point, k1 is the curvature of the path point, δ is the wheel angle, is the rate of change of the horizontal coordinate at the path point in the Frenet coordinate system, Δu represents the rate of change of the control quantity, u min and u max are the maximum and minimum values ​​of the control variables, respectively; l(x,u), g(x,u), and f(x,u) represent functions of x and u.

2. The method for tracking trajectory of a reconfigurable distributed drive multi-axis vehicle according to claim 1, characterized in that: In the step (3): The lateral force of the wheel is expressed by the following formula: T is the coefficient matrix. When the axis i exists, t i is a positive integer 1, otherwise it is 0; the expression of α is as follows: Explain the symbols and numbers in the above formula: α is the slip angle of the wheel, d is the wheelbase, and k is the tire cornering stiffness; The longitudinal dynamics model of the vehicle is as follows: Among them, F x is the sum of the longitudinal forces acting on the vehicle in the vehicle body coordinate system, C D Indicates the drag coefficient, A f is the frontal area of ​​the vehicle, ρ0 is the air density, and d1 is the interference term.

3. The reconfigurable distributed drive multi-axis vehicle trajectory tracking method according to claim 2, characterized in that: The step (4) comprises: The velocity tracking error is defined as: yes v =v x -v xd Introduce the integral term into the sliding mode function: Define the Lyapunov function as: In order to satisfy the Lyapunov stability condition, that is, The control rate of the longitudinal force of the designed vehicle is Explain the symbols and numbers in the above formula: c, k2, ε, D are all positive constants, v xd is the target longitudinal velocity, v x is the actual longitudinal velocity, e v is the difference between the actual longitudinal velocity and the desired velocity, s is the sliding mode surface function, and in order to alleviate the chattering phenomenon that is easily produced by ordinary sliding mode controllers, the hyperbolic tangent function tanh is used instead of the ordinary switching function:

4. A transport device comprising a vehicle body, characterized in that: Also includes: Sensors are used to obtain the longitudinal and lateral motion speeds, yaw angular velocity and position information of the vehicle body in real time; The controller controls the steering system and wheel motors of the vehicle body according to the detection information of the sensor to execute the trajectory tracking method of the reconfigurable distributed drive multi-axle vehicle according to any one of claims 1 to 3.

Images