Hierarchical control method for three-axis modular distributed electric-drive heavy-load vehicle

Abstract

The invention discloses a hierarchical control method for a three-axis modular distributed electric drive heavy-duty vehicle. Establishing a two-degree-of-freedom vehicle dynamics model considering the longitudinal force and the lateral force of the tire; the input quantity of the two-degree-of-freedom model is a three-axis front wheel turning angle, the vehicle yaw velocity in a steady state is solved in combination with the longitudinal velocity of the vehicle, and meanwhile driver pedal input is analyzed into the target longitudinal velocity of the vehicle; setting two control variables of total driving force and total yaw moment, and tracking a reference longitudinal speed and a yaw velocity issued by an upper layer by using a model predictive control algorithm; and a cost function is set by taking the tracking error of the total driving force and the total yawing moment, the tire adhesion rate and the energy economy as soft constraints, a multi-objective optimization problem is established by taking the maximum moment and the minimum moment which can be output by the hub motor as hard constraints, solving is performed by adopting quadratic programming, and finally the control quantity is issued. According to the method, pure drive-by-wire control of the three-axle six-wheel vehicle is achieved by means of the modern control theory, and application and development of heavy-load vehicles are facilitated.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent chassis control, and in particular relates to a hierarchical control method for a three-axle modular distributed electric drive heavy-load vehicle. Background Art

[0002] The modular distributed electric drive heavy-duty vehicle adopts a new variable configuration chassis, which has the advantages of flexible configuration and strong scalability. The new configuration chassis contains multiple connectable driving modules, each of which is composed of several composite functional units with independent drive, braking, steering and suspension functions, and has the advantages of flexible configuration and strong scalability. The above-mentioned configuration chassis adopts an independently controllable wheel hub motor drive system, a wire-controlled braking system, a wire-controlled steering system and a wire-controlled suspension system, which doubles the number of controllable actuators and increases the degree of control freedom. The composite functional unit structure adopts the form of a vehicle. When turning, the wheels have no fixed geometric constraints. The maximum turning angle of a single wheel can be increased from ±35° to ±60°, which can effectively reduce the vehicle's turning radius. The new variable configuration chassis decouples mechanical constraints, increases the degree of control freedom, and improves the vehicle's flexibility and maneuverability through coordinated control of each wheel. It can effectively broaden the driving performance boundaries of heavy-duty vehicles, thereby supporting safe and efficient heavy-duty transportation.

[0003] Modular distributed electric drive heavy-duty vehicles bring new challenges and problems while improving the performance boundaries of vehicles. The use of all-wheel independent steering / drive / braking leads to mechanical decoupling of various functional mechanisms and underconstraint of lateral movement of wheels. The actuation of multiple actuators is strongly coupled with the "longitudinal-lateral-vertical" motion state of the vehicle, and the coordination of multi-wheel independent steering is difficult. The first task to solve the above problems is to develop a basic control architecture as the cornerstone, and on this basis, improve the control theory to enhance the high-speed driving stability of heavy-duty vehicles and realize safe and efficient transportation of heavy-duty vehicles. Therefore, the hierarchical control method of three-axis modular distributed electric drive heavy-duty vehicles mentioned in the present invention has important application value. Summary of the Invention

[0004] Purpose of the invention: The present invention provides a hierarchical control method for a three-axle modular distributed electric drive heavy-load vehicle, which can be applied to the actual control application of the three-axle modular distributed electric drive heavy-load vehicle, and serve as the basis for the simulation development and hardware-in-the-loop simulation of the three-axle modular distributed electric drive heavy-load vehicle.

[0005] Technical solution: The three-axle modular distributed electric drive heavy-duty vehicle described in the present invention includes the following steps:

[0006] (1) Driving Intention Analysis: A two-degree-of-freedom vehicle dynamics model is established that takes into account tire longitudinal and lateral forces. The input to the two-degree-of-freedom model is the front wheel angles along all three axes. Combined with the vehicle's longitudinal velocity, the vehicle's yaw rate in steady state is calculated. The driver's pedal input is then converted into the vehicle's target longitudinal velocity.

[0007] (2) Vehicle state tracking: Two control quantities, total driving force and total yaw moment, are established, and the reference longitudinal velocity and yaw angular velocity sent down by the upper layer are tracked using the model predictive control algorithm.

[0008] (3) Drive torque distribution: The cost function is set using the tracking error of the total drive force and total yaw torque, as well as tire adhesion and energy economy as soft constraints. The maximum and minimum torques that the hub motor can output are used as hard constraints. A multi-objective optimization problem is established and solved using quadratic programming. The control variables are finally issued.

[0009] Furthermore, the implementation process of step (1) is as follows:

[0010] The three-axle steering angle is determined by a steering multi-model set, which includes center and rear axle counter-steering, front wheel steering, center and rear axle same-direction steering, and diagonal steering.

[0011] The driver's steering wheel input angle controls the front axle angle, while the center and rear axle angles are determined by the driver's steering mode and the driver's set angle ratio coefficient between the center and rear axles and the front wheels. The relationship is as follows:

[0012]

[0013] Where: m is the steering mode, Δ is the angle ratio coefficient between the center rear axle and the front wheel.

[0014] Finally, we get δ f , δ m , δ r , directly sent to the chassis for steering control and serves as the input for the subsequent two-degree-of-freedom model.

[0015] The pedal depth input by the driver is then analyzed. The input accelerator pedal depth percentage is considered as the maximum acceleration percentage, and the desired vehicle speed is obtained by integrating it. The equation is as follows:

[0016] v x_des =∫(P d ·a max )dt

[0017] In the two-degree-of-freedom model, this paper focuses on only one state variable, the yaw angular velocity. Therefore, a model is built for the yaw direction, taking into account both the longitudinal and lateral forces of the tires. The vehicle force diagram is shown in the figure. The following equation is obtained:

[0018]

[0019] Where: i = f, m, r represent the front axle, middle axle and rear axle respectively; j = l, r represent the left and right wheels respectively; F ijx Indicates the longitudinal force on each tire; Fijy Indicates the lateral force on each tire; l i It represents the distance between the three axles and the center of mass, which is positive in front of the center of mass and negative behind the center of mass; b represents the half wheelbase, which is the same for the three axles.

[0020] The modular distributed electric drive vehicle uses a hub motor, which can directly obtain the driving force of each wheel. After modeling the tire rolling resistance, F ijx ; F ijy Obtained by estimation.

[0021] Let the above formula You can get ω r The steady-state value ω r_des .

[0022] Therefore, the expected values of vehicle speed and yaw rate v can be obtained x_des and ω r_des .

[0023] The implementation process of step (2) is as follows:

[0024] The system model is designed as the following state space equation:

[0025]

[0026] The tracking state is the expected value v of the vehicle speed and yaw rate mentioned above. x_des and ω r_des , the control input is the total driving torque T total -T roll -T wind and the total yaw moment M z_total The constraints are set as the sum of the maximum positive and negative torques that the motor can output and the maximum and minimum values of the vehicle speed and yaw rate expected when the algorithm is applied.

[0027] Rolling resistance torque T roll Modeling is based on the linear rolling resistance torque model. The specific formula is as follows:

[0028] T roll = rmgf(c1+c2v x )

[0029] Where: r is the tire radius, m is the vehicle mass, g is the acceleration of gravity, f is the road condition coefficient, c1 is the rolling resistance constant coefficient, and c2 is the rolling resistance speed correlation coefficient.

[0030] The air resistance moment is calculated according to the air resistance formula. The specific formula is as follows:

[0031]

[0032] Where: r is the tire radius, ρ is the air density, C d is the air resistance coefficient, A is the frontal area, and v is the vehicle speed.

[0033] Then, an MPC model predictive control algorithm is established, and the cost is set as tracking error and control increment. The cost function is as follows:

[0034]

[0035] Where: Np is the prediction interval, S is the terminal cost weight, Q is the state weight, and R is the control weight.

[0036] Convert it into a quadratic programming problem and solve it to get the control quantity T total -T roll -T wind and M z_total Control quantity T total -T roll -T wind Plus T roll 、T wind Get T total , and get the final control quantity T total and M z_total .

[0037] The implementation process of step (3) is as follows:

[0038] Step (3) is to satisfy M z_total Under the premise of T total Assigned to each wheel. Therefore, the cost function is set as follows:

[0039] J=α1·(TT total )+α2·(M z -M z_total )+α3·φ+α4·ψ

[0040] α is the weight coefficient of each cost, is the adhesion rate, is the energy economy, and η is the motor efficiency.

[0041] Constraints can be set based on the requirements of the final application, such as the minimum and maximum torque output of each wheel and the maximum output power of the motor. The optimization problem is then transformed into a quadratic programming problem and solved. The resulting torque distribution plan for all six wheels is then transmitted to the chassis for control.

[0042] Beneficial Effects: Compared with existing technologies, this invention has the following advantages: 1. It is suitable for use in six-wheeled, independent, modular, distributed electric drive vehicles, enabling full process control from driver input to final torque distribution under multiple constraints. 2. This invention can serve as a foundation for guiding the development and scientific research of more advanced functions for six-wheeled, independent, modular, distributed electric drive vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of a steering multi-model set according to an embodiment of the present invention;

[0045] Figure 3 2 is a force diagram of a vehicle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The present invention will be further described in detail below with reference to the accompanying drawings.

[0047] See Figure 1 (1) Driving intention analysis: A two-degree-of-freedom vehicle dynamics model is established that takes into account the longitudinal and lateral forces of the tires. The input of the two-degree-of-freedom model is the front wheel angle of the three axes. Combined with the vehicle longitudinal velocity, the vehicle yaw rate in the steady state is solved. At the same time, the driver's pedal input is analyzed as the vehicle's target longitudinal velocity.

[0048] (2) Vehicle state tracking: Two control quantities, total driving force and total yaw moment, are established, and the reference longitudinal velocity and yaw angular velocity sent down by the upper layer are tracked using the model predictive control algorithm.

[0049] (3) Drive torque distribution: The cost function is set using the tracking error of the total drive force and total yaw torque, as well as tire adhesion and energy economy as soft constraints. The maximum and minimum torques that the hub motor can output are used as hard constraints. A multi-objective optimization problem is established and solved using quadratic programming. The control variables are finally issued.

[0050] Furthermore, the implementation process of step (1) is as follows:

[0051] like Figure 2 The three-axle steering angle is determined by the steering multi-model set. The steering multi-model set includes center and rear axle counter-steering, front wheel steering, center and rear axle same-direction steering, and diagonal steering.

[0052] The driver's steering wheel input angle controls the front axle angle, while the center and rear axle angles are determined by the driver's steering mode and the driver's set angle ratio coefficient between the center and rear axles and the front wheels. The relationship is as follows:

[0053]

[0054] Where: m is the steering mode, Δ is the angle ratio coefficient between the center rear axle and the front wheel.

[0055] Finally, we get δ f , δ m , δ r , directly sent to the chassis for steering control and serves as the input for the subsequent two-degree-of-freedom model.

[0056] The pedal depth input by the driver is then analyzed. The input accelerator pedal depth percentage is considered as the maximum acceleration percentage, and the desired vehicle speed is obtained by integrating it. The equation is as follows:

[0057] v x_des =∫(P d ·a max )dt

[0058] In the two-degree-of-freedom model, the present invention only focuses on the yaw angular velocity as a state variable. Therefore, the yaw direction is modeled, taking both the longitudinal force and the lateral force of the tire into account. The vehicle force diagram is as follows: Figure 3 As shown. The following equation is obtained:

[0059]

[0060] Where: i = f, m, r represent the front axle, middle axle and rear axle respectively; j = l, r represent the left and right wheels respectively; F ijx Indicates the longitudinal force on each tire; F ijy Indicates the lateral force on each tire; l i It represents the distance between the three axles and the center of mass, which is positive in front of the center of mass and negative behind the center of mass; b represents the half wheelbase, which is the same for the three axles.

[0061] The modular distributed electric drive vehicle uses a hub motor, which can directly obtain the driving force of each wheel. After modeling the tire rolling resistance, F ijx ; F ijy Obtained by estimation.

[0062] Let the above formula You can get ω r The steady-state value ω r_des .

[0063] Therefore, the expected values of vehicle speed and yaw rate v can be obtained x_des and ω r_des .

[0064] The task of step (2) is to track the expected value of vehicle speed and yaw rate v x_des and ω r_des The system model is designed as the following state space equation:

[0065]

[0066] The tracking state is the expected value v of the vehicle speed and yaw rate mentioned above. x_des and ω r_des , the control input is the total driving torque T total -T roll -T wind and the total yaw moment M z_total The constraints are set as the sum of the maximum positive and negative torques that the motor can output and the maximum and minimum values of the vehicle speed and yaw rate expected when the algorithm is applied.

[0067] Rolling resistance torque T roll Modeling is based on the linear rolling resistance torque model. The specific formula is as follows:

[0068] T roll = rmgf(c1+c2v x )

[0069] Where: r is the tire radius, m is the vehicle mass, g is the acceleration of gravity, f is the road condition coefficient, c1 is the rolling resistance constant coefficient, and c2 is the rolling resistance speed correlation coefficient.

[0070] The air resistance moment is calculated according to the air resistance formula. The specific formula is as follows:

[0071]

[0072] Where: r is the tire radius, ρ is the air density, C d is the air resistance coefficient, A is the frontal area, and v is the vehicle speed.

[0073] Then, an MPC model predictive control algorithm is established, and the cost is set as tracking error and control increment. The cost function is as follows:

[0074]

[0075] Where: Np is the prediction interval, S is the terminal cost weight, Q is the state weight, and R is the control weight.

[0076] Convert it into a quadratic programming problem and solve it to get the control quantity T total -T roll -T wind and M z_total Control quantity T total -T roll -T wind Plus T roll 、T wind Get T total , and get the final control quantity T total and M z_total .

[0077] Step (3) is to satisfy M z_total Under the premise of T total Assigned to each wheel. Therefore, the cost function is set as follows:

[0078] J=α1·(TT total )+α2·(M z -M z_total )+α3·φ+α4·ψ

[0079] α is the weight coefficient of each cost, is the adhesion rate, is the energy economy, and η is the motor efficiency.

[0080] Constraints can be set based on the requirements of the final application, such as the minimum and maximum torque output of each wheel and the maximum output power of the motor. The optimization problem is then transformed into a quadratic programming problem and solved. The resulting torque distribution plan for all six wheels is then transmitted to the chassis for control.

[0081] 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.

[0082] 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.

[0083] The term “connection” as used in this application may mean a direct connection between components or an indirect connection between components via other components.

[0084] 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 hierarchical control method for a three-axle modular distributed electric drive heavy-duty vehicle, characterized in that: The following steps are involved: (1) A two-degree-of-freedom vehicle dynamics model is established that takes into account tire longitudinal and lateral forces. The input of the two-degree-of-freedom model is the front wheel angle of the three axes. Combined with the vehicle longitudinal velocity, the vehicle yaw rate in the steady state is solved. At the same time, the driver's pedal input is resolved into the vehicle's target longitudinal velocity. (2) Establish two control variables: total driving force and total yaw moment, and use the model predictive control algorithm to track the reference longitudinal velocity and yaw angular velocity sent from the upper layer; (3) The cost function is set with the tracking error of the total driving force and total yaw moment, as well as the tire adhesion rate and energy economy as soft constraints. The maximum and minimum torque that the hub motor can output are used as hard constraints. A multi-objective optimization problem is established and solved using quadratic programming. Finally, the control quantity is issued.

2. A hierarchical control method for a three-axle modular distributed electric drive heavy-duty vehicle according to claim 1, characterized in that: The implementation process of step (1) is as follows: The three-axle steering angle is determined by the steering multi-model set, which includes center and rear axle opposite-direction steering, front wheel steering, center and rear axle same-direction steering, and diagonal steering; The driver's steering wheel input angle controls the front axle angle. The center and rear axle angles are determined by the driver's steering mode and the driver-set ratio coefficient of the center and rear axles to the front wheels. The relationship is as follows: Where: m is the steering mode, Δ is the angle ratio coefficient between the rear axle and the front wheel; Get the steering angle δ of the three axes f , δ m , δ r , sent to the chassis for steering control, and also serves as input for the subsequent two-degree-of-freedom model; The pedal depth input by the driver is analyzed, and the input accelerator pedal depth percentage is regarded as the maximum acceleration percentage. The expected vehicle speed is obtained by integrating it. The equation is as follows: in x_des =∫(P d ·and max )dt(2) Where: v x_des is the expected vehicle speed, P d is the pedal depth percentage entered by the driver, a max is the maximum acceleration of the vehicle; Modeling the lateral direction of the vehicle, taking both the longitudinal and lateral forces of the tires into account, yields the following equation: Where: i = f, m, r represent the front axle, middle axle and rear axle respectively; j = l, r represent the left and right wheels respectively; F ijx Indicates the longitudinal force on each tire; F ijy Indicates the lateral force on each tire; l i represents the distance between the three axles and the center of mass, which is positive in front of the center of mass and negative behind the center of mass; b represents the half wheelbase, which is the same for the three axles; δ ij Represents the rotation angle of each wheel; ω r is the vehicle yaw angular velocity, v x is the longitudinal velocity of the vehicle; v y is the lateral speed of the vehicle, M z_total is the sum of the yaw moments caused by the tire forces on the six wheels, m represents the mass of the vehicle, ω r is the vehicle yaw angular velocity; The modular distributed electric drive vehicle uses a hub motor to obtain the driving force of each wheel and the tire rolling resistance is modeled to obtain F ijx ; Let the above formula Get ω r The steady-state value ω r_des ; Get the expected value v of vehicle speed and yaw rate x_des and ω r_des .

3. A hierarchical control method for a three-axle modular distributed electric drive heavy-duty vehicle according to claim 2, characterized in that: The implementation process of step (2) is as follows: The system model is designed as the following state space equation: The tracking state is the expected value v of the vehicle speed and yaw rate mentioned above. x_des and ω r_des , the control input is the total longitudinal torque T total -T roll -T wind and the total yaw moment M z_total , set the constraints to the sum of the maximum positive and negative torques that the motor can output and the maximum and minimum values of the vehicle speed and yaw rate expected when the algorithm is applied; Rolling resistance torque T roll Modeling is based on the linear rolling resistance torque model. The specific formula is as follows: T roll =rmgf(c1+c2v x ) Where: r is the tire radius, m is the vehicle mass, g is the acceleration of gravity, f is the road condition coefficient, c1 is the rolling resistance constant coefficient, c2 is the rolling resistance speed correlation coefficient; v x is the vehicle speed; The air resistance moment is calculated according to the air resistance formula. The specific formula is as follows: Where: r is the tire radius, ρ is the air density, C d is the air resistance coefficient, A is the frontal area, and v is the vehicle speed; An MPC model predictive control algorithm is established, with the cost set as tracking error and control increment. The cost function is as follows: Where: Np is the prediction interval, S is the terminal cost weight, Q is the state weight, and R is the control weight; is the prediction of the tracking error at time k to time k+Np, (Δu) [k+i|k] is the control increment at time k+Np calculated at time k; Convert it into a quadratic programming problem and solve it to get the control quantity T total -T roll -T wind and M z_total , control quantity T total -T roll -T wind Plus T roll 、T wind Get T total , and get the final control quantity T total and M z_total .

4. A hierarchical control method for a three-axle modular distributed electric drive heavy-duty vehicle according to claim 3, characterized in that: The implementation process of step (3) is as follows: Set the cost function as follows: J=α1·(T ij cosδ ij -T total )+α2·(M z -M z_total )+α3·φ+α4·ψ α1, α2, and α3 are the weight coefficients of each cost, is the adhesion rate, is the energy economy, η is the motor efficiency, Mz is the sum of the total yaw moments caused by the lateral and longitudinal tire forces on the six vehicles, as shown in formula (3), M z_total is the sum of the calculated yaw moment control variables, T ij is the driving torque of the six wheels, T total is the sum of the calculated torque control quantities, δ ij is the turning angle of the six wheels; Constraints are set according to the requirements of the final application, and the optimization problem is converted into a quadratic programming problem for solution. Finally, the torque distribution plan for the six wheels is obtained and sent to the chassis for control.

Images