Eight-wheel hub motor driving vehicle matrix vector running control technology

Abstract

This invention proposes a matrix vector driving control method for electric motor-driven wheeled vehicles. The method includes: constructing a planar motion model of the electric motor-driven wheeled vehicle, the motion model outputting a desired yaw rate and sideslip angle; designing a vehicle controller based on the planar motion model; wherein the vehicle controller includes an upper controller and a lower controller; the upper controller is a yaw torque controller, which includes: a sideslip angle control module, which accepts the sideslip angle as input and outputs the torque required for sideslip angle control; and a yaw rate control module, which accepts the sideslip angle as input and outputs the torque required for yaw rate control; the torque required for sideslip angle control and the torque required for yaw rate control are controlled by a coordination control module to obtain a final output control quantity; the final output control quantity is input to the lower controller for torque vector control and slip rate control, thereby outputting the torque required for each wheel of the electric motor-driven wheeled vehicle, and thus controlling the electric motor-driven wheeled vehicle.

Description

Technical Field

[0001] This invention relates to matrix vector driving control technology for eight-wheel hub motor driven vehicles, and more specifically, to a matrix vector driving control method for motor-driven wheeled vehicles. Background Technology

[0002] Multi-wheeled off-road vehicles or vehicles used for mobile missions, such as 8x8 distributed electric drive wheeled armored vehicles, employ an all-wheel independent drive system based on hub motors. While meeting vehicle driving requirements, the torque of each wheel can be arbitrarily distributed within the performance range of the drive motors, achieving different torque distribution methods between axles and wheels. Compared to traditional internal combustion engine wheeled vehicles, each drive motor in an 8x8 distributed electric drive wheeled armored vehicle can be independently controlled without affecting others. Therefore, it is easier to achieve functions such as anti-lock braking, anti-slip braking, yaw moment control, and rollover prevention, and it has unique advantages in enhancing vehicle stability and improving overall vehicle energy efficiency. However, this also places higher demands on the vehicle's driving control strategy, namely, how to achieve the vehicle's driving control objectives by rationally distributing the torque of multiple hub motors under different road conditions, operating conditions, and usage requirements, to ensure the vehicle's driving stability under complex operating conditions.

[0003] Existing technologies have conducted considerable research on how to fully leverage the advantages of distributed drive. One common method is direct yaw moment control, which involves tracking the yaw rate and sideslip angle to obtain the overall yaw moment required for stable vehicle operation and distributing it to each wheel. This control method faces two main challenges in its application to electrically driven armored vehicles: first, how the upper-level controller coordinates the control of the yaw rate and sideslip angle; and second, how to effectively distribute torque from the entire vehicle to each wheel. Other optimization methods exist, such as multivariable control methods based on yaw rate tracking, parallel control methods based on single-input single-output yaw rate and sideslip angle, and torque distribution methods using tire utilization as the optimization objective. While these methods attempt to improve vehicle handling and stability, they struggle to meet the complex road conditions and driving requirements of wheeled off-road or armored vehicles, thus limiting their practical application.

[0004] The present invention aims to solve the above problems by designing a control system and control method that meet the needs of complex road conditions and multi-task requirements, so as to meet the application requirements of wheeled electric drive vehicles. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a matrix vector driving control method for motor-driven wheeled vehicles. The method includes: constructing a planar motion model of the motor-driven wheeled vehicle, the motion model outputting a desired yaw rate and sideslip angle; designing a vehicle controller based on the planar motion model; wherein the vehicle controller includes an upper-level controller and a lower-level controller; the upper-level controller is a yaw torque controller, comprising: a sideslip angle control module, which accepts the sideslip angle as input and outputs the torque required for sideslip angle control; and a yaw rate control module, which accepts the sideslip angle as input and outputs the torque required for yaw rate control; the torque required for sideslip angle control and the torque required for yaw rate control are controlled by a coordination control module to obtain a final output control quantity; the final output control quantity is input to the lower-level controller for torque vector control and slip rate control, thereby outputting the torque required for each wheel of the motor-driven wheeled vehicle, and thus controlling the motor-driven wheeled vehicle.

[0006] Preferably, the lower-level controller includes a torque vector control module and a slip ratio control module. The torque vector control module takes the output of the yaw moment controller as input and outputs the motor torque of the drive wheel. The motor torque of the drive wheel is corrected by the slip mode control module.

[0007] Preferably, the yaw rate control module includes a yaw rate sliding film controller, the sliding film surface of which is:

[0008]

[0009] In the formula: s γ c1 and c2 are the sliding mode variables for yaw rate control; c1 and c2 are the weights of the yaw rate error and its rate of change, respectively, c1 > 0, c2 > 0, β is the sideslip angle of the center of mass, and γ is the yaw rate. d For the desired yaw rate, β d The desired centroid sideslip angle;

[0010]

[0011] In the formula, P is an intermediate variable.

[0012]

[0013] make Where ε1 and k1 are parameters of the exponential reaching law, both greater than 0, and sgn is the sign function; therefore, we have,

[0014]

[0015] Integrating the above equation, we obtain the yaw torque ΔM required for yaw rate control. γ .

[0016] Preferably, the centroid sideslip angle control module includes a centroid sideslip angle sliding film controller, the sliding film surface of which is:

[0017]

[0018] In the formula: c3 and c4 are the weights of the sideslip angle error and its rate of change, respectively; ΔM is the yaw moment control quantity of the sideslip angle. β for

[0019]

[0020] In the formula: ε2, k2, exponential reaching law parameters; Q is an intermediate variable.

[0021]

[0022] Preferably, the coordinated control module includes a weighting function G, which adjusts the yaw rate and the sideslip angle. After adjustment, the final output control quantity ΔM of the yaw moment controller is... z The size is

[0023] ΔM z =G·ΔM γ +(1-G)ΔM β

[0024] The weighting function is:

[0025]

[0026] In the formula: a and b are the threshold values ​​for controlling the center of gravity side slip angle; when the center of gravity side slip angle β≤a, only the yaw rate is controlled; when the center of gravity side slip angle a<β≤b, the two are jointly controlled, and the larger the center of gravity side slip angle, the greater its weight; when the center of gravity side slip angle β>b, only the center of gravity side slip angle is controlled; and the values ​​of the threshold values ​​a and b can be used to divide the control area according to the relationship between the center of gravity side slip angle and the tire lateral force, and then the final boundary values ​​are determined through simulation experiments.

[0027] Preferably, it further includes: the final output control quantity ΔM of the yaw moment controller. z The torque vector method is used for synthesis, and the synthesis steps include the synthesis of total driving force and the synthesis of yaw moment; the wheel torques of the motor-driven wheeled vehicle are divided into four groups according to the axle, then the longitudinal force F of the vehicle is obtained. xd and yaw moment M z The size is:

[0028]

[0029] In the formula: F xij (i = 1, 2, 3, 4 represent the i-th axis, j = 1, 2 represent the left and right sides) represents the longitudinal force of the wheel; δ xij (i = 1, 2 are the i-th axis, j = 1, 2 are the left and right sides) is the wheel angle; d is the wheelbase.

[0030] Preferably, the lower-level controller includes a torque vector control layer, in which torque vector control and slip ratio control are performed. The torque vector control includes total driving force synthesis and yaw moment synthesis.

[0031] Preferably, the total driving force synthesis is based on the driver's needs and the vertical load ratio of the wheels to synthesize the total driving force for the vehicle's forward movement.

[0032] Preferably, the yaw moment synthesis includes: converting ΔM obtained from the yaw moment controller... z The torque vectors of the first two axles are proportionally combined, while the torque vectors of the last two axles are not included in the yaw moment synthesis. The additional yaw moment is proportionally synthesized between the axles, with equal additions and subtractions at the two wheels on the same axle to maintain a constant total driving force.

[0033]

[0034] In the formula: ΔFxij (i=1,2 is the i-th axis, j=1,2 is the left and right sides) is the longitudinal force adjustment of the wheel;

[0035] The final driving force can be obtained by adding the driving forces of the eight wheels and the yaw moment adjustment amounts separately.

[0036] Preferably, the slip ratio control includes single-wheel drive anti-slip control and coaxial torque adjustment; the purpose of the anti-slip control is to reduce the drive torque setpoint; and the actual slip ratio s is compared with the desired slip ratio s. d When the deviation e is large, the given torque should be reduced rapidly; at s and s d When the deviation is small, a certain amount of steady-state error is eliminated.

[0037] Preferably, the anti-slip control uses a PPI controller; P control is used when the error is large, and PI control is used when the error is small. Attached Figure Description

[0038] Various embodiments or examples (“Examples”) of this disclosure are disclosed in the following detailed description and accompanying drawings. It is not necessary to draw the drawings to scale. Generally, unless otherwise specified in the claims, the operations of the methods disclosed in this invention can be performed in any order. In the drawings:

[0039] Figure 1 A planar motion model of a distributed electric drive vehicle according to the present invention is shown;

[0040] Figure 2 A vehicle control system for a distributed electric drive vehicle according to the present invention is shown;

[0041] Figure 3 The relationship between the yaw efficiency and turning radius of each drive wheel on each axle;

[0042] Figure 4 A structural diagram of the torque vector controller designed according to the present invention;

[0043] Figure 5 This is a schematic diagram of the slip ratio controller according to the present invention;

[0044] Figure 6 This is a real-time simulation experimental platform used in this invention;

[0045] Figure 7 The experimental results are from a simulation experiment of obstacle avoidance in a wheeled vehicle using the control system of the present invention.

[0046] Figure 8 The results are from a simulation experiment on anti-skid performance of a wheeled vehicle using the control system of this invention. Detailed Implementation

[0047] Before explaining one or more embodiments of this disclosure in detail, it should be understood that the embodiments are not limited to the construction details in their specific applications, and the steps or methods presented in the following embodiments or drawings. The systems and methods of the present invention will now be described in detail with reference to the accompanying drawings.

[0048] I. Constructing a planar motion model of a wheeled vehicle

[0049] Assuming the vehicle operates in a linear region with a constant longitudinal velocity, and the lateral force of the tires is proportional to the slip angle, neglecting the influence of rolling resistance. Ignoring the effects of the steering and suspension systems, and treating the vehicle's longitudinal velocity as constant, we only consider lateral and yaw motion, establishing a linear two-degree-of-freedom vehicle planar motion model, as follows: Figure 1 As shown. Figure 1 In the text, v represents the vehicle speed. x v y These are the vehicle's longitudinal speed and lateral speed, respectively, F xi F yi (i = 1, 2, 3, 4 represent the i-th axis) are the longitudinal force and lateral force of the wheel, respectively, L i α represents the distance from the i-th axis to the vehicle's center of mass. i β is the wheel slip angle, γ is the center of mass slip angle, γ is the yaw rate, δ1 and δ2 are the wheel steering angles of the first and second axles, O is the vehicle center of mass, and O′ is the vehicle steering center.

[0050] The relationship between a vehicle's lateral acceleration and lateral force satisfies the following:

[0051]

[0052] The yaw acceleration and yaw moment of a vehicle satisfy the following relationship:

[0053]

[0054] Where: m is the total vehicle mass; I z Let be the moment of inertia about the center of mass.

[0055] From equations (1) and (2), the desired yaw rate γ can be obtained. d And the centroid side slip angle β d for

[0056]

[0057]

[0058] In the formula: C i These are the lateral stiffnesses of the tires on each axle, a s The ratio of the steering angles of the front two axles; μ is the ground adhesion coefficient; g is the acceleration due to gravity;

[0059] L is the equivalent wheelbase.

[0060]

[0061] K is the stability coefficient.

[0062]

[0063] II. Design of the vehicle controller

[0064] 1. Determine the vehicle control method

[0065] For vehicle driving control systems, the controller is the core component. The controller analyzes driver input signals and vehicle motion states, and calculates the torque commands for each drive motor based on the corresponding control strategy. The controller designed in this paper consists of two layers: an upper-layer yaw torque controller and a lower-layer torque vector controller, as shown in the diagram. Figure 2 As shown. Figure 2 In the middle, ΔM β The yaw moment required to control the sideslip angle of the center of gravity, ΔM γ The yaw torque required for yaw rate control, ΔM zThe total yaw moment is obtained through weighted calculation. First, the driver provides a control signal, and the vehicle's two-degree-of-freedom model obtains the desired yaw rate and sideslip angle based on the control signal. Second, the upper-level yaw moment controller calculates the corresponding yaw moments and coordinates them according to weighting coefficients. Finally, the lower-level torque vector controller distributes the yaw moment according to the torque vector synthesis method. Simultaneously, to prevent excessive wheel torque and slippage, a slip rate controller limits the wheel output torque.

[0066] 2. Design of yaw moment controller

[0067] The function of the yaw moment controller is to determine the required yaw moment for the vehicle during operation. Sliding mode control is characterized by good robustness, rapid response, simple control, and applicability to nonlinear systems; therefore, this invention employs a sliding mode control algorithm. After adding yaw moment control to the two-degree-of-freedom model, its differential equation is:

[0068]

[0069] Design a yaw rate sliding mode controller, defining the sliding surface as...

[0070]

[0071] In the formula: s γ c1 and c2 are the sliding mode variables for yaw rate control; c1 and c2 are the weights of the yaw rate error and its rate of change, respectively, where c1 > 0 and c2 > 0.

[0072] Substituting equation (7) into equation (8), we get

[0073]

[0074] In the formula, P is an intermediate variable.

[0075]

[0076] Design a sliding mode control law using an exponential reaching law, so that... Where ε1 and k1 are parameters of the exponential reaching law, both greater than 0, and sgn is the sign function. Typically, severe chattering occurs during rapid reaching. To reduce chattering, k1 should be appropriately increased while ε1 should be appropriately decreased. Combining equation (9), we get...

[0077]

[0078] Integrating equation (11), we obtain the yaw torque ΔM required for yaw rate control. γ .

[0079] Design a sliding mode controller for the centroid side slip angle, defining the sliding surface as...

[0080]

[0081] In the formula, c3 and c4 are the weights of the sideslip angle error and its rate of change, respectively. Referring to the yaw rate control, a control law is designed using an exponential reaching law to obtain the yaw moment control quantity ΔM for the sideslip angle. β for

[0082]

[0083] In the formula: ε2, k2, exponential reaching law parameters; Q is an intermediate variable.

[0084]

[0085] There is a complex coupling relationship between yaw rate and sideslip angle. This invention uses a weighted function G to coordinate the control of these two factors. When the vehicle's sideslip angle is small, the gain between yaw rate and steering wheel angle can characterize the vehicle's stability; in this case, controlling only the yaw rate can keep the vehicle stable. However, when the vehicle experiences fishtailing or sideslipping, the sideslip angle becomes large and increases rapidly. Since fishtailing and sideslipping have a relatively small impact on the yaw rate, the yaw rate cannot accurately reflect the vehicle's operating state under these conditions, and sideslip angle control is necessary to effectively control the vehicle's motion. Therefore, a weighted function is used to adjust both factors, and the final output control quantity ΔM of the yaw torque controller is... z The size is

[0086] ΔM z =G·ΔM γ +(1-G)ΔM β (15)

[0087] To make the transition process smoother, the weighting function G is designed using the Z membership function (zmf), and its expression is as follows:

[0088]

[0089] In the formula, a and b are the threshold values ​​for control using the center-of-gravity sideslip angle. When the center-of-gravity sideslip angle is less than β ≤ a, only the yaw rate is controlled; when the center-of-gravity sideslip angle a < β ≤ b, both are controlled jointly, with a larger weight given to the larger side-of-gravity angle; when the center-of-gravity sideslip angle β > b, only the center-of-gravity sideslip angle is controlled. The values ​​of the thresholds a and b can be determined by dividing the control region according to the relationship between the center-of-gravity sideslip angle and the tire lateral force, and then determining the final boundary values ​​through simulation experiments.

[0090] 3. Design a torque vector controller

[0091] (1) Perform yaw performance analysis

[0092] ΔM obtained from the yaw moment controller z The torque vector control method is used for synthesis. The wheel torques are divided into four groups according to the axle. Since the thrust generated by the wheels is parallel to the ground, the resultant force can only be parallel to the ground; therefore, only the vehicle's motion in a plane is considered. The vehicle's longitudinal force F... xd and yaw moment M z The size is

[0093]

[0094] In the formula: F xij (i = 1, 2, 3, 4 represent the i-th axis, j = 1, 2 represent the left and right sides) represents the longitudinal force of the wheel; δ xij (i = 1, 2 are the i-th axis, j = 1, 2 are the left and right sides) is the wheel angle; d is the wheelbase.

[0095] The above equations are underdetermined, with infinitely many solutions, meaning there are infinitely many torque combinations that can achieve the control objective. To find a particular solution, two approaches are needed: first, determine the allocation rules, increasing the number of equations; second, use optimization methods, setting optimization objectives and constraints to minimize the cost of achieving the control objective. This invention employs a rule-based allocation method. For a simple yaw torque allocation problem, the simplest approach is the average allocation principle, where the increase or decrease in driving torque on both the left and right drive wheels is the same. This method is simple and direct, but it fails to consider the differences in yaw action of the drive wheels. Yaw torque is essentially the result of tire force. Due to factors such as vehicle structural characteristics and steering wheel angle, different wheels will produce different yaw torques when the same driving or braking torque is applied. This paper defines the ratio of the yaw torque generated by the drive wheel to the driving torque of that drive wheel as the yaw efficiency. Therefore, the yaw efficiency K of drive wheel i is... i It can be represented as

[0096]

[0097] In the formula: T i T is the output torque of drive wheel i. i =F i ·r,F i ΔM is the longitudinal force of the tire, r is the effective radius of the tire; i For F i The resulting yaw torque around the center of the vehicle.

[0098] The tire force F resulting from the additional yaw torque applied to the tire is considered to act directly along the tire's longitudinal direction. Clearly, when the same force F is applied to the non-steering wheels of the two rear axles (i=3, i=4), the resulting yaw torque is the same, with a magnitude of...

[0099]

[0100] like Figure 1 The two-degree-of-freedom linear model of the vehicle shown, taking left turn as an example, when the steering wheels of the front two axles (i=1, i=2) are subjected to the same force F, the resulting yaw torques are respectively

[0101]

[0102] Therefore, the relationship between the steering wheel angle and the steering radius can be approximately expressed as follows:

[0103]

[0104] In the formula: R is the turning radius.

[0105] Combining equations (18) to (21), we can obtain

[0106]

[0107] The relationship between its turning radius and the yaw efficiency of each drive wheel is as follows: Figure 3 As shown.

[0108] Depend on Figure 3 It can be seen that when the turning radius is small, the yaw efficiency of the front two axles (steering axles) is high, and the difference is greater as the turning radius becomes smaller. When distributing yaw torque, increasing the yaw torque component of the wheels with high yaw torque efficiency can improve the driving force utilization efficiency and increase the dynamic response speed. When the turning radius of the vehicle increases, its yaw angular velocity decreases, and the required yaw torque is smaller. At this time, using the front two axles to synthesize the yaw torque can meet the steering needs. Therefore, the rear two axles only need to provide the longitudinal force required for the vehicle to move forward. Therefore, the additional yaw torque is only synthesized by the torque vectors of the steering wheels, that is, the first two sets of vectors, and the latter two sets of vectors only participate in the synthesis of longitudinal force.

[0109] (2) Design a torque vector controller

[0110] Figure 4 The diagram shown is a structural diagram of a torque vector controller designed according to the present invention. Figure 4 In the diagram, T1 to T8 represent the torques of the eight wheels, and T′... 1-8 To synthesize the torque of the eight wheels required for the longitudinal driving force, ΔT 1-8 The torque adjustment for the first four wheels required to generate the additional yaw moment. Figure 4 It can be seen that the torque vector control process is mainly divided into the following two steps: the first step is to synthesize the total driving force, and the second step is to synthesize the yaw moment to obtain the motor torque of the eight drive wheels.

[0111] First, the total driving force for vehicle forward movement is synthesized based on the driver's needs and the vertical load ratio of the wheels. Since the vertical load of the wheels is a dynamically changing quantity during vehicle movement, this invention calculates it in two parts: static load and dynamic change. Factors causing dynamic load changes mainly include vehicle acceleration / deceleration, steering, climbing, and obstacle crossing. This paper only considers the vehicle's motion in a plane, ignoring factors such as climbing and obstacle crossing, and disregarding its roll and pitch motion. It only analyzes the load changes when the vehicle is traveling on a horizontal road, assuming that its load is only affected by longitudinal / lateral acceleration. Therefore, the vertical load of the eight wheels is expressed as follows:

[0112]

[0113] In the formula: F zij (i = 1, 2, 3, 4 represent the i-th axis, j = 1, 2 represent the left and right sides) represents the vertical load on the wheel; h represents the height of the vehicle's center of gravity above the ground; a x a y L represents the vehicle's longitudinal and lateral accelerations. a L b As an intermediate variable,

[0114] L a =(L1-L2)+(L1+L3)+(L1+L4) (24)

[0115] L b 2 =(L1-L2) 2 +(L1+L3) 2 +(L1+L4) 2 (25)

[0116] The total driving force requirement of the vehicle is synthesized according to the vertical load ratio, and the calculation formula is as follows:

[0117]

[0118] Secondly, the ΔM obtained from the yaw moment controller z The torque vectors of the first two axles are proportionally combined, while the torque vectors of the latter two axles do not participate in the yaw moment synthesis. Since the turning angle of the wheels on the first axle is greater than that on the second axle, and there is a fixed ratio between them, the additional yaw moment is proportionally combined between the axles. The addition and subtraction of the yaw moment are the same for the two wheels on the same axle to maintain a constant total driving force.

[0119]

[0120] In the formula: ΔFxij (i=1,2 is the i-th axis, j=1,2 is the left and right sides) is the longitudinal force adjustment of the wheel.

[0121] Equations (26) and (27) provide the driving force of each wheel that meets the longitudinal force requirement and the adjustment amount of the driving force of the first four wheels that meets the yaw moment requirement. By adding the driving forces of the eight wheels and the adjustment amount of the yaw moment respectively, the final driving force can be obtained.

[0122] (3) Design of slip ratio controller

[0123] Because the driving force of a vehicle is limited by the maximum coefficient of adhesion on the ground, if the applied driving force is too large, the wheels will slip significantly, causing the lateral force to approach saturation and the vehicle's lateral stability to decrease rapidly. Therefore, it is essential to control the slip ratio during wheel drive. Traditional mechanical vehicles typically control wheel slip ratio through traction control systems or differential locks. In a hub motor-driven armored vehicle, each wheel is driven independently, forming a relatively independent system that allows for individual wheel control. During anti-slip control, the primary function is to reduce the given driving torque. When the deviation between the actual slip ratio *s* and the desired slip ratio *sd* is large (e), the given torque must be rapidly reduced, resulting in a larger anti-slip control output value. When the deviation between *s* and *sd* is small, a certain amount of steady-state error needs to be eliminated. P-control is used when the error is large, and PI control is used when the error is small. Figure 5 The diagram shown is a schematic of a P-PI segmented controller.

[0124] When some wheels slip during a vehicle's turn, the anti-slip controller reduces the torque output of the target motor. If the torque of other drive motors is not corrected, the overall vehicle yaw torque output will inevitably be affected, differing from the expected value, and the desired stable vehicle steering cannot be achieved.

[0125] From the perspective of maintaining the balance between the vehicle's longitudinal driving force and yaw moment, there are two methods: First, maintain the overall longitudinal driving force of the vehicle constant. In this case, the driving torque (ΔT) lost by the slipping wheel can be evenly distributed to the other non-slipping wheels on the same side. This method will cause changes in the yaw moment synthesized by the torque vector between groups, which may not meet the total yaw moment requirement. Second, maintain the total yaw moment constant. In this case, the torque of the other drive wheel on the same axle as the slipping wheel can also be reduced by ΔT. The advantage is that the yaw moment synthesized by the torque vector between groups remains unchanged, but the overall driving force of the vehicle is reduced. Considering that the main reason for vehicle slippage is the low road adhesion coefficient, and the driving force of each wheel is close to the adhesion limit, reducing the torque can improve the vehicle's lateral force margin, thereby improving the vehicle's lateral stability and handling stability. Therefore, the control method of maintaining a constant yaw moment is adopted.

[0126] III. Example Verification

[0127] This invention employs real-time simulation experiments based on dSPACE software to verify the proposed control system and control method. The hardware in the real-time simulation platform structure is as follows: Figure 6 As shown, this platform implements a human-vehicle closed-loop feedback system consisting of a driver-controller (dSPACE), a hub motor model (RT-Lab), and a vehicle model (Vortex).

[0128] The functions of each module in the hardware-in-the-loop real-time simulation platform are as follows:

[0129] 1) Driver control system. The driver can control the vehicle by using the accelerator pedal, brake pedal, steering wheel, and gear shift, and the control signals are transmitted to the dSPACE control terminal via the CAN bus.

[0130] 2) Controller System. Composed of the dSPACE real-time simulation system, it implements the vehicle control algorithm based on the driver's operating commands and vehicle status, and sends control signals to the drive motor via the bus.

[0131] 3) Motor drive system. A real-time simulation model of the hub motor is run in RT-Lab, receiving control commands from the controller and sending the operating status to the controller and the Vortex vehicle dynamics model.

[0132] 4) Real-time dynamics simulation system. The Vortex simulation system performs real-time simulation and scene display of vehicle dynamics, obtains the real-time operating status of the vehicle, and feeds it back to other systems. Before the simulation experiment, the accuracy of the vehicle model has been verified by comparison with real vehicle tests and can meet the simulation requirements.

[0133] The simulation object is an 8×8 distributed electric drive wheeled armored vehicle. The main parameters of the vehicle and the motor are shown in Table 1. The motor type is a permanent magnet synchronous motor.

[0134] Table 1 Main parameters of the vehicle and motor

[0135]

[0136]

[0137] 1. Good road surface obstacle avoidance driving simulation

[0138] Referring to the double lane change test method, the driving process of emergency obstacle avoidance by a vehicle was simulated to test some of the vehicle's dynamic performance and handling stability. The simulation conditions were: road adhesion coefficient of 0.85, initial vehicle speed of 60 km / h, time of 12 s, distance of 200 m, and a control group was set up with a vehicle with 8 wheels having even torque distribution and no yaw moment control. Figure 7 The simulation results of this experiment are shown below.

[0139] according to Figure 7The simulation results show that the driver first manipulates the steering wheel and pedals to make the vehicle travel along a predetermined trajectory. From... Figure 7 Simulation results show that the steering wheel angle begins to change at 3 seconds. Without control, the lateral deviation of the uncontrolled vehicle's trajectory from the ideal trajectory at the point of maximum lateral displacement is 2.4 m, the maximum yaw rate is 15.4° / s, and the maximum sideslip angle is 2.3°. After applying yaw moment control, the wheel torque is redistributed, creating a torque difference between the inner and outer wheels, thus generating yaw moment. This ensures that the torque on the outer wheel is greater than that on the inner wheel during turning. At the point of maximum lateral displacement, the lateral deviation of the trajectory from the ideal trajectory decreases to 0.5 m, and the yaw rate and sideslip angle significantly decrease with smoother changes. These results demonstrate that yaw moment and torque vector control effectively control the vehicle, enabling it to travel along a predetermined trajectory and improving its driving stability.

[0140] 2. Simulation of acceleration driving on split-road surfaces

[0141] To verify the effectiveness of the anti-skid controller when the adhesion conditions of the wheels on both sides are different, a driving simulation on a split road was conducted, with the road adhesion coefficients set to 0.3 and 0.85 on both sides respectively. During the simulation, the driver did not operate the steering wheel, but only pressed the accelerator pedal. The vehicle's running state under no-control and anti-skid control conditions was recorded, focusing on the wheels on both sides of the first axle. Figure 8 As shown.

[0142] Depend on Figure 8 (a) and Figure 8 (c) It can be seen that in the uncontrolled state, due to the different road surface adhesion systems on both sides, under the same given torque, the wheel speed on the low-adhesion side is significantly higher than that on the high-adhesion side, and the wheel slips; from 8(b) and Figure 8 (d) It can be seen that by applying anti-skid control, the slippage of the wheel on the low-adhesion side is significantly controlled; Figure 8 (e) It can be seen that, under uncontrolled conditions, the torque value given by the accelerator pedal is distributed to the left and right motors. The actual output value of the motor on the high-adhesion side is basically consistent with the given value, while on the low-adhesion side, due to wheel slippage causing excessive motor speed, the actual output torque of the motor is significantly lower than the value given by the accelerator pedal. Figure 8 Comparing (e) and 8(f), it can be seen that under anti-slip control, the torque of the wheel on the low-adhesion side is adjusted, which improves the wheel slippage state. At the same time, the torque of the wheel on the high-adhesion side is also adjusted, so that the total yaw moment is zero, maintaining the lateral stability of the vehicle.

[0143] In summary, this invention proposes a direct yaw moment and torque vector control method for an 8×8 distributed electric drive wheeled armored vehicle. In implementing this invention, a hierarchical controller was designed. The upper-level controller uses sliding mode control to calculate the yaw moment based on the yaw rate and sideslip angle, and designs a weighted function to coordinate the output of the total yaw moment. The lower-level controller divides the eight wheels into four vector groups according to their axles. Based on the yaw moment and longitudinal force requirements, the torque of each wheel is obtained using a torque vector synthesis method, and an anti-slip controller is designed to control the wheel slip rate. This invention also verifies the proposed control method using hardware on a real-time simulation platform. Experimental results show that under the proposed control strategy, the vehicle exhibits better response characteristics compared to an uncontrolled vehicle; the actual yaw rate and trajectory track the expected values ​​well, improving the vehicle's driving stability.

[0144] Although the invention has been described with reference to embodiments shown in the accompanying drawings, equivalent or alternative means may be used without departing from the scope of the claims. The components described and illustrated in this invention are merely examples of systems / apparatus and methods that can be used to implement embodiments of this disclosure, and may be replaced with other devices and components without departing from the scope of the claims.

Claims

1. A matrix vector driving control method for an electric motor-driven wheeled vehicle, comprising: Construct a planar motion model of the motor-driven wheeled vehicle, and the motion model outputs the desired yaw rate and center-of-gravity sideslip angle; Based on the aforementioned planar motion model, a vehicle controller is designed; characterized in that... The vehicle controller includes an upper-level controller and a lower-level controller; The upper-level controller is a yaw moment controller, which includes: The center of mass sideslip angle control module accepts the center of mass sideslip angle as input and outputs the torque required to control the center of mass sideslip angle. And a yaw rate control module, which accepts the center of mass sideslip angle as input and outputs the torque required for yaw rate control; The torque required for controlling the centroid side slip angle and the torque required for controlling the yaw rate are obtained as the final output control quantity after being controlled by the coordination control module; The final output control quantity is input to the lower-level controller to perform torque vector control and slip ratio control, thereby outputting the torque required for each wheel of the motor-driven wheeled vehicle, and thus controlling the motor-driven wheeled vehicle. The coordinated control module includes a weighting function G, which adjusts the yaw rate and the sideslip angle. After adjustment, the final output control quantity ΔM of the yaw moment controller is... z The size is: ΔM z =G·ΔM γ +(1-G)ΔM β Where γ is the yaw rate, β is the sideslip angle of the center of mass, and ΔM γ The yaw torque ΔM required for yaw rate control β This is the yaw moment control quantity for the sideslip angle of the center of gravity; The weighting function G is: In the formula: a and b are the threshold values ​​for controlling the sideslip angle; when the sideslip angle β≤a, only the yaw rate is controlled; when the sideslip angle a<β≤b, both are controlled together, and the larger the sideslip angle, the greater its weight; when the sideslip angle β>b, only the sideslip angle is controlled; and the values ​​of the threshold values ​​a and b can be used to divide the control area according to the relationship between the sideslip angle and the tire lateral force, and then the final boundary values ​​are determined through simulation experiments.

2. The matrix vector driving control method according to claim 1, characterized in that, The lower-level controller includes a torque vector control module and a slip ratio control module. The torque vector control module takes the output of the yaw moment controller as input and outputs the motor torque of the drive wheel. The motor torque of the drive wheel is corrected by the slip ratio control module.

3. The matrix vector driving control method according to claim 1, characterized in that, The yaw rate control module includes a yaw rate sliding film controller, whose sliding film surface is: In the formula: s γ c1 and c2 are the sliding mode variables for yaw rate control; c1 and c2 are the weights of the yaw rate error and its rate of change, respectively, c1 > 0, c2 > 0, and γ is the yaw rate. d The desired yaw rate; In the formula, I z Let P be the moment of inertia about the center of mass, and let P be an intermediate variable. In the formula, m is the total vehicle mass; v x C represents the longitudinal speed of the vehicle. i These represent the lateral stiffness of the tires on each axle, L. i This represents the distance of the i-th axis from the vehicle's center of gravity, where i = 1, 2, 3, 4 represents the i-th axis; C1 and C3 are the lateral stiffness of the tires on the 1st and 3rd axes, respectively; L1 and L2 represent the distances of the 1st and 2nd axes from the vehicle's center of gravity, respectively; a s δ1 is the wheel steering angle ratio of the first two axles; δ1 is the wheel steering angle of the first axle. make Where ε1 and k1 are parameters of the exponential reaching law, both greater than 0, and sgn is the sign function; therefore, we have, Integrating the above equation, we obtain the yaw torque ΔM required for yaw rate control. γ .

4. The matrix vector driving control method according to claim 1, characterized in that, The centroid sideslip angle control module includes a centroid sideslip angle sliding film controller, whose sliding film surface is: In the formula: c3 and c4 are the weights of the centroid sideslip angle error and its rate of change, respectively; β is the centroid sideslip angle, β d The desired sideslip angle; the yaw moment control quantity ΔM for the sideslip angle. β for In the formula: ε2, k2, exponential reaching law parameters; Q is an intermediate variable. In the formula: γ is the yaw rate; I z v is the moment of inertia about the center of mass; m is the total mass of the vehicle; v x C represents the longitudinal speed of the vehicle. i These represent the lateral stiffness of the tires on each axle, L. i This represents the distance of the i-th axis from the vehicle's center of gravity, where i = 1, 2, 3, 4 represents the i-th axis. C1 and C3 are the lateral stiffness of the tires on the 1st and 3rd axes, respectively. L1 and L2 represent the distances of the 1st and 2nd axes from the vehicle's center of gravity, respectively. s δ1 is the ratio of the wheel angles of the first two axles; δ1 is the wheel steering angle of the first axle.

5. The matrix vector driving control method according to claim 1, characterized in that, Also includes: The final output control quantity ΔM of the yaw moment controller z The torque vector method is used for synthesis, which includes the synthesis of total driving force and yaw moment; the wheel torques of the motor-driven wheeled vehicle are divided into four groups according to the axle, then the longitudinal force F of the vehicle is obtained. xd and yaw moment M z The size is: In the formula: F xi F represents the longitudinal force of the wheel, where i = 1, 2, 3, 4 represent the i-th axis; xij The longitudinal force of the wheel is represented by i = 1, 2, 3, 4, which represents the i-th axis, and j = 1, 2, which represents the left and right sides; δ ij Let δ1 and δ2 be the wheel steering angles, i = 1 and 2 be the i-th axle, j = 1 and 2 be the left and right sides, respectively; δ1 and δ2 are the wheel steering angles of the first and second axles, respectively; δ3 and δ4 are the wheel steering angles of the third and fourth axles, respectively; d is the wheelbase; L1 and L2 represent the distances of the first and second axles from the vehicle's center of gravity, respectively.

6. The matrix vector driving control method according to claim 1, characterized in that, The lower-level controller includes a torque vector control layer, in which torque vector control and slip ratio control are performed. The torque vector control includes the synthesis of total driving force and the synthesis of yaw moment.

7. The matrix vector driving control method according to claim 6, characterized in that, The total driving force synthesis is based on the driver's needs and the vertical load ratio of the wheels to synthesize the total driving force for the vehicle's forward movement.

8. The matrix vector driving control method according to claim 7, characterized in that, The yaw moment synthesis includes: converting ΔM obtained from the yaw moment controller... z The torque vectors of the first two axles are proportionally combined, while the torque vectors of the last two axles do not participate in the yaw moment synthesis. The additional yaw moment is proportionally synthesized between the axles, with equal additions and subtractions at the two wheels on the same axle to maintain a constant total driving force. In the formula: ΔFxij is the longitudinal force adjustment of the wheel, i = 1, 2 is the i-th axle, j = 1, 2 is the left and right sides; δ ij Let be the wheel turning angle, i = 1, 2 be the i-th axle, j = 1, 2 be the left and right sides; L1 and L2 represent the distances of the first and second axles from the vehicle's center of gravity, respectively; d is the wheelbase. The final driving force can be obtained by adding the driving forces of the eight wheels and the yaw moment adjustment amounts separately.

9. The matrix vector driving control method according to claim 6, characterized in that, The slip ratio control includes single-wheel drive anti-slip control and coaxial torque adjustment; the purpose of the anti-slip control is to reduce the drive torque setpoint; and to balance the actual slip ratio s with the desired slip ratio s. d When the deviation e is large, the given torque should be reduced rapidly; at s and s d When the deviation is small, a certain amount of steady-state error is eliminated.

10. The matrix vector driving control method according to claim 9, characterized in that, The anti-slip control uses a PPI controller; P control is used when the error is large, and PI control is used when the error is small.

Images