Distributed electric drive unmanned vehicle torque distribution control method and system with optimized energy consumption

Abstract

The invention relates to an energy-consumption-optimized torque distribution control method and system for a distributed electrically-driven unmanned vehicle, belongs to the technical field of automatic driving, and solves the problems that in an existing distributed driving system, path tracking precision and energy consumption efficiency are difficult to consider at the same time, multi-motor cooperative control is complex, and the tire adhesion utilization rate is unbalanced. The method includes: acquiring a current driving state of an unmanned vehicle; obtaining a required traction torque and an equivalent required steering angle of the unmanned vehicle based on the current driving state and the expected driving state of the unmanned vehicle; on the basis of the equivalent demand steering angle and the transverse stability constraint, the needed direct yawing moment is obtained; taking the required traction torque and the required vehicle speed as indexes, and selecting a current optimal driving mode from a pre-constructed optimal driving mode partition table; and in the optimal driving mode, based on the required direct yawing moment, obtaining an output torque distribution result of each driving motor and distributing the output torque distribution result to the corresponding driving wheel.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a torque distribution control method and system for a distributed electric drive unmanned vehicle with optimized energy consumption. Background Art

[0002] With the rapid development of electric vehicle technology, especially in the field of driverless vehicle control technology, the demand for improving energy efficiency and vehicle performance is becoming increasingly urgent. In practical applications, distributed drive electric platforms are highly favored due to their excellent maneuverability and obstacle navigation capabilities in complex terrains.

[0003] However, distributed drive systems also bring new challenges. First, how to effectively manage energy while maintaining vehicle stability and improving maneuverability has become a key issue. Traditional centralized electric drive platforms usually involve only one or two motors, while distributed drive systems require independent control of multiple motors, which increases the complexity of control. In addition, existing torque distribution strategies often do not fully consider the efficiency area of ​​the motor, resulting in inefficient energy utilization.

[0004] Direct yaw moment control (DYC) is a method to improve vehicle stability and handling by controlling the vehicle's yaw moment. However, traditional DYC systems usually only contain two layers of control structure, the upper layer calculates the required external lateral torque, and the lower layer distributes the torque between the wheels. This method is not accurate enough when dealing with complex off-road environments, which limits its application in high-performance ground unmanned platforms.

[0005] In terms of path tracking accuracy and energy efficient management, existing control strategies often have difficulty meeting both requirements. On the one hand, in order to improve path tracking accuracy, the torque of each wheel needs to be precisely controlled; on the other hand, in order to optimize energy efficiency, the torque needs to be reasonably distributed to ensure that the motor operates in a high-efficiency area. Summary of the invention

[0006] In view of the above analysis, the embodiments of the present invention aim to provide a torque distribution control method and system for a distributed electric drive unmanned vehicle with optimized energy consumption, so as to solve the problems in existing distributed drive systems such as the difficulty in balancing path tracking accuracy and energy efficiency, the complexity of multi-motor coordinated control, and the uneven utilization of tire adhesion.

[0007] The purpose of the present invention is mainly achieved through the following technical solutions:

[0008] The present invention provides a torque distribution control method for a distributed electric drive unmanned vehicle with optimized energy consumption, comprising the following steps:

[0009] Get the current driving status of the unmanned vehicle;

[0010] Based on the current driving state and the expected driving state of the unmanned vehicle, obtaining the required traction torque and the equivalent required steering angle of the unmanned vehicle;

[0011] Based on the equivalent required steering angle and the lateral stability constraint of the unmanned vehicle, obtaining a direct yaw moment required by the unmanned vehicle;

[0012] Using the required traction torque and the required vehicle speed as indexes, selecting the current optimal driving mode of the unmanned vehicle from a pre-constructed optimal driving mode partition table;

[0013] In the optimal driving mode, based on the direct yaw moment required by the unmanned vehicle, the output torque distribution result of each driving motor is obtained and distributed to the corresponding driving wheels.

[0014] Furthermore, the driving state includes a driving speed and a driving speed azimuth;

[0015] The step of obtaining the required traction torque and the equivalent required steering angle of the unmanned vehicle includes:

[0016] Based on the current driving speed and the expected driving speed of the unmanned vehicle, a speed tracking error of the unmanned vehicle is obtained;

[0017] Based on the speed tracking error, obtaining a required traction torque of the unmanned vehicle;

[0018] Obtaining a path tracking error of the unmanned vehicle based on the current driving speed azimuth of the unmanned vehicle and the tangent azimuth of the road at the preview point;

[0019] Based on the path tracking error, an equivalent required steering angle of the unmanned vehicle is obtained.

[0020] Furthermore, obtaining the path tracking error of the unmanned vehicle includes:

[0021] Based on the current driving speed azimuth of the unmanned vehicle and the tangent azimuth of the road at the preview point, an equivalent reference fusion deviation angle of the unmanned vehicle is obtained;

[0022] Based on the current position of the unmanned vehicle and the position of the preview point, obtaining a lateral position deviation of the unmanned vehicle;

[0023] Based on the equivalent reference fusion deviation angle and the lateral position deviation, obtaining a calculated value of an equivalent fusion deviation angle of the unmanned vehicle;

[0024] Based on the equivalent reference fusion deviation angle and the calculated equivalent fusion deviation angle, a path tracking error of the unmanned vehicle is obtained.

[0025] Further, based on the speed tracking error, PI control is used to obtain the required traction torque of the unmanned vehicle;

[0026] Based on the path tracking error, PI control is used to obtain the equivalent required steering angle of the unmanned vehicle.

[0027] Furthermore, obtaining the direct yaw moment required by the unmanned vehicle based on the equivalent required steering angle and the lateral stability constraint of the unmanned vehicle includes:

[0028] Based on the lateral stability constraint of the unmanned vehicle, a maximum direct yaw moment intervention value is obtained;

[0029] The intervention degree of the yaw moment is solved using a high-order fitting curve, and the direct yaw moment required by the unmanned vehicle is obtained using the following formula:

[0030] M zd =sgn(δ d )(-2Δ 3 +3Δ 2 )M dmax

[0031] Δ=abs(δ d / δ max )

[0032] Among them, M zd represents the direct yaw moment; δ d represents the equivalent required steering angle; δ max represents the maximum mechanical steering angle of the unmanned vehicle; M dmax represents the maximum direct yaw moment intervention value; abs() represents the absolute value function; sgn() represents the sign function.

[0033] Furthermore, the optimal driving mode includes a combination of driving motors involved in driving;

[0034] The step of obtaining the output torque distribution result of each driving motor includes:

[0035] Based on the required traction torque, direct yaw moment and torque output values ​​of each driving motor involved in the driving of the unmanned vehicle, an objective function and corresponding constraint conditions are constructed with the goal of minimizing the weighted sum of torque distribution error and tire load;

[0036] Based on the constraint conditions, the objective function is solved using a quadratic programming algorithm to obtain an output torque distribution result of each driving motor involved in the driving.

[0037] Furthermore, the objective function is expressed as:

[0038]

[0039] u=S·u′

[0040] v d =[T d M zd ] T

[0041] Wherein, B represents the efficiency matrix of the unmanned vehicle; u represents the output torque distribution vector of each driving motor involved in the driving of the unmanned vehicle; S represents the driving mode matrix of the unmanned vehicle; u′ represents the torque output value of all driving motors of the unmanned vehicle; v d represents the control output demand expectation matrix; W1 represents the weight matrix of torque distribution error; W2 represents the weight matrix of tire load; T d represents the required traction torque of the unmanned vehicle; M zd represents the direct yaw moment; δ d represents the equivalent required steering angle of the unmanned vehicle.

[0042] Furthermore, the constraint condition is:

[0043] The resultant force of the longitudinal force and the lateral force of each tire involved in driving the unmanned vehicle does not exceed the adhesion limit of the tire;

[0044] The output torque of each wheel hub motor driving the unmanned vehicle does not exceed the adhesion limit of the tire;

[0045] The output torque and torque change rate of each wheel hub motor driven by the unmanned vehicle do not exceed the physical limits of each motor.

[0046] Furthermore, the optimal drive mode partition table is pre-constructed by the following steps:

[0047] Obtaining various driving modes and different driving conditions of the unmanned vehicle;

[0048] For each driving condition, calculating the energy consumption efficiency of the unmanned vehicle in each driving mode and normalizing each driving condition;

[0049] For each driving condition, the driving mode with the highest energy efficiency is selected as the optimal driving mode, and an optimal driving mode partition table with each driving condition as an index is obtained.

[0050] A torque distribution control system for a distributed electric drive unmanned vehicle with optimized energy consumption, comprising:

[0051] A tracking error calculation module, used to obtain a speed tracking error and a path tracking error of the unmanned vehicle based on a current driving state and an expected driving state of the unmanned vehicle;

[0052] A lateral / longitudinal tracking control module, configured to obtain a required traction torque and an equivalent required steering angle of the unmanned vehicle based on a speed tracking error and a path tracking error obtained by the tracking error calculation module;

[0053] A compound steering equivalent steering angle matching calculation module, used to obtain the direct yaw moment required by the unmanned vehicle based on the equivalent required steering angle obtained by the lateral / longitudinal tracking control module and the lateral stability constraint of the unmanned vehicle;

[0054] A driving mode selection module, configured to select the current optimal driving mode of the unmanned vehicle from a pre-constructed optimal driving mode partition table using the required traction torque and required vehicle speed obtained by the lateral / longitudinal tracking control module as indexes;

[0055] The torque distribution module is used to obtain the output torque distribution result of each drive motor based on the direct yaw torque required by the unmanned vehicle in the optimal driving mode and distribute it to the corresponding drive wheels.

[0056] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0057] 1. The present invention realizes the integrated control of path tracking and energy management through a hierarchical control framework. The upper-level controller is responsible for determining the required total driving torque and yaw moment, the middle-level controller controls the torque distribution between axles according to the energy consumption, and the lower-level controller refines the torque distribution of each driving wheel based on vehicle stability and path tracking performance, thereby realizing effective decoupling of energy management and torque distribution, which can not only improve the path tracking performance of the unmanned platform, but also realize efficient energy management.

[0058] 2. The present invention calculates the energy consumption efficiency distribution diagram under different driving modes offline through bench testing, and then determines the optimal motor operating area through the speed and required torque index, ensuring that the motor operates under high-efficiency conditions and can ensure that the torque is distributed to maximize the energy conversion efficiency.

[0059] 3. The present invention takes into account factors such as tire vertical load distribution and path tracking error, and adopts a quadratic programming algorithm to achieve output torque distribution of each drive motor as well as steering angle and braking torque, which can effectively improve the driving control stability of the unmanned platform off-road maneuverability.

[0060] 4. The present invention introduces a fault-tolerant control mechanism, which monitors the fault status of each drive motor in real time and recalculates the control allocation according to the failure factor matrix to ensure that the stability and reliability of the vehicle can be maintained in the event of a motor failure, and has higher adaptability and robustness.

[0061] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like components throughout the drawings.

[0063] Figure 1 A schematic flow chart of a torque distribution control method for a distributed electric drive unmanned vehicle with energy consumption optimization in an embodiment of the present invention;

[0064] Figure 2 Schematic diagram of the efficiency distribution of the optimal driving mode in an embodiment of the present invention;

[0065] Figure 3 The schematic diagram is a structural diagram of a torque distribution control system for a distributed electric drive unmanned vehicle with optimized energy consumption in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0067] A specific embodiment of the present invention discloses a torque distribution control method for a distributed electric drive unmanned vehicle with optimized energy consumption, such as Figure 1 As shown, the steps S1 to S5 are included:

[0068] Step S1, obtaining the current driving state of the unmanned vehicle, wherein the driving state includes the driving speed and the driving speed azimuth.

[0069] Specifically, based on the sensors carried by the unmanned vehicle, the current driving speed and driving speed azimuth of the unmanned vehicle are obtained. Exemplarily, the sensors include a global navigation satellite system (GPS), an inertial measurement unit (IMU) and a wheel speed sensor.

[0070] The driving speed is the current longitudinal speed of the unmanned vehicle.

[0071] The driving speed azimuth is the angle between the current driving direction of the unmanned vehicle and the reference direction.

[0072] Step S2: based on the current driving state and the expected driving state of the unmanned vehicle, obtain the required traction torque and the equivalent required steering angle of the unmanned vehicle.

[0073] Specifically, the path tracking control goal of the unmanned vehicle is to make the vehicle drive along the preset path while maintaining the desired speed and direction. Therefore, according to the difference between the current driving state and the desired driving state, the required traction torque and steering angle are calculated to adjust the vehicle's driving state to make it as close to the desired state as possible.

[0074] Furthermore, obtaining the required traction torque and the equivalent required steering angle of the unmanned vehicle includes steps S21 to S24:

[0075] Step S21: Based on the current driving speed and the expected driving speed of the unmanned vehicle, the speed tracking error δ of the unmanned vehicle is obtained using the following formula: v :

[0076] δ v =v xref -v x

[0077] Among them, v xref represents the expected driving speed; v x Indicates the current speed of the driverless car.

[0078] Step S22: obtaining the required traction torque of the unmanned vehicle based on the speed tracking error.

[0079] Specifically, the required traction torque of the unmanned vehicle is used to adjust the longitudinal speed of the vehicle to reach a desired speed. If the current speed is lower than the desired speed, the required traction torque will increase to accelerate the vehicle; otherwise, the required traction torque will decrease to decelerate the vehicle.

[0080] Furthermore, based on the speed tracking error, PI control is used to obtain the required traction torque of the unmanned vehicle.

[0081] Specifically, the PI controller is a linear controller that adjusts the control amount according to the size of the deviation and the integral of the deviation. In this embodiment, the speed tracking error is input into the PI controller, and the required traction torque is calculated by integrating the proportional and integral parts to adjust the speed of the unmanned vehicle. The formula is as follows:

[0082] T d =K pv δ v +K iv ∫δ v

[0083] Among them, T d represents the required traction torque of the unmanned vehicle; Kpv Represents the proportional gain coefficient; K iv Represents the integral gain coefficient.

[0084] Step S23: obtaining a path tracking error of the unmanned vehicle based on the current driving speed azimuth of the unmanned vehicle and the tangent azimuth of the road at the preview point.

[0085] Specifically, the preview point represents the next point that the unmanned vehicle will reach during path planning; the preview point road tangent azimuth is an angle obtained by the on-board integrated navigation system carried by the unmanned vehicle, indicating the ideal heading that the unmanned vehicle should maintain at the preview point.

[0086] Furthermore, obtaining the path tracking error of the unmanned vehicle includes steps S231 to S234:

[0087] Step S231: based on the current driving speed azimuth of the unmanned vehicle and the tangent azimuth of the road at the preview point, obtain the equivalent reference fusion deviation angle of the unmanned vehicle.

[0088] Specifically, the equivalent reference fusion deviation angle of the unmanned vehicle represents the difference between the azimuth of the road tangent at the preview point and the azimuth of the current driving speed of the unmanned vehicle, which is used to describe the deviation between the current driving direction of the unmanned vehicle and the expected driving direction, that is, the heading angle error. The heading angle error is obtained using the following formula:

[0089] θ eref =θ ref -θ vehicle

[0090] Among them, θ eref represents the equivalent reference fusion deviation angle; θ ref θ represents the tangent azimuth of the road at the preview point; vehicle Indicates the current driving speed azimuth of the unmanned vehicle.

[0091] Step S232: Based on the current position of the unmanned vehicle and the position of the preview point, obtain the lateral position deviation of the unmanned vehicle.

[0092] Specifically, the lateral position deviation of the unmanned vehicle is used to represent the vertical distance between the current position of the unmanned vehicle and the preset path. The lateral position deviation is obtained using the following formula:

[0093] d err =(x ref -x vehicle )sin(θ ref )-(y ref -y vehicle )cos(θ ref )

[0094] Among them, derr Indicates the lateral position deviation; (x vehicle ,y vehicle ) represents the current position of the unmanned vehicle; (x ref ,y ref ) represents the position of the preview point.

[0095] Step S233: obtaining a calculated value of an equivalent fusion deviation angle of the unmanned vehicle based on the equivalent reference fusion deviation angle and the lateral position deviation.

[0096] Specifically, the equivalent fusion deviation angle calculation value is an indicator that combines the heading angle error and the lateral position deviation, and is used to more comprehensively describe the deviation between the current driving state of the unmanned vehicle and the expected driving state. The equivalent fusion deviation angle calculation value is obtained using the following formula:

[0097] θ e =θ eref +k·d err

[0098] Among them, θ e represents the calculated value of the equivalent fusion deviation angle; k represents the proportional coefficient, which is obtained based on the current speed of the unmanned vehicle and the curvature of the path, and is used to adjust the influence of the lateral position deviation on the calculated value of the equivalent fusion deviation angle. For example, 0.1 can be selected.

[0099] Step S234: obtaining the path tracking error of the unmanned vehicle based on the equivalent reference fusion deviation angle and the calculated value of the equivalent fusion deviation angle.

[0100] Specifically, the path tracking error represents the comprehensive deviation of the direction and position of the unmanned vehicle, and the path tracking error is obtained using the following formula:

[0101] δ θ =θ eref -θ e

[0102] Among them, δ θ represents the path tracking error.

[0103] Step S24: obtaining an equivalent required steering angle of the unmanned vehicle based on the path tracking error.

[0104] Specifically, the equivalent required steering angle of the unmanned vehicle is used to adjust the steering angle of the vehicle so that it travels along a preset path.

[0105] Furthermore, based on the path tracking error, PI control is used to obtain the equivalent required steering angle of the unmanned vehicle.

[0106] In this embodiment, the path tracking error is input into the PI controller, and the equivalent required steering angle is calculated by integrating the proportional and integral parts to adjust the direction of travel of the unmanned vehicle. The formula is as follows:

[0107] δ d =K pθ δ θ +K iθ ∫δ θ

[0108] Among them, δ d represents the equivalent required steering angle of the unmanned vehicle; K pθ Represents the proportional gain coefficient; K iθ Represents the integral gain coefficient.

[0109] Step S3: obtaining a direct yaw moment required by the unmanned vehicle based on the equivalent required steering angle and the lateral stability constraint of the unmanned vehicle.

[0110] Specifically, the lateral stability constraint of the unmanned vehicle is a restriction condition imposed on the stability of the unmanned vehicle in the lateral direction, that is, perpendicular to the driving direction of the vehicle, to ensure that the vehicle remains stable during driving.

[0111] In the path tracking control of the unmanned vehicle, due to the understeering characteristics caused by the rear wheel steering, that is, the actual steering response of the vehicle lags behind the expected steering command, it is difficult to achieve flexible and accurate steering by adjusting the steering angle alone. Therefore, when solving the equivalent required steering angle δ d After that, the required direct yaw moment needs to be further calculated.

[0112] The direct yaw moment is to enhance the steering response of the vehicle by applying an additional yaw moment, thereby more effectively compensating for understeering characteristics and making the vehicle more flexible and stable when turning.

[0113] Furthermore, obtaining the direct yaw moment required by the unmanned vehicle based on the equivalent required steering angle and the lateral stability constraint of the unmanned vehicle includes:

[0114] Based on the lateral stability constraint of the unmanned vehicle, a maximum direct yaw moment intervention value is obtained.

[0115] Specifically, the maximum direct yaw moment intervention value is the maximum yaw moment that can be safely applied by the vehicle, calculated according to the lateral stability constraint of the vehicle, during the path tracking control and dynamic control of the unmanned vehicle. This value can ensure that the vehicle will not lose stability due to excessive yaw moment during driving, thereby ensuring the safety and handling performance of the vehicle. In this embodiment, the maximum direct yaw moment intervention value is determined based on simulation or actual vehicle experiments.

[0116] Furthermore, the intervention degree of the yaw moment is solved using a high-order fitting curve, and the direct yaw moment required by the unmanned vehicle is obtained using the following formula:

[0117] M zd =sgn(δ d )(-2Δ 3 +3Δ 2 )M dmax

[0118] Δ=abs(δ d / δ max )

[0119] Among them, M zd represents the direct yaw moment; δ d represents the equivalent required steering angle; δ max represents the maximum mechanical steering angle of the unmanned vehicle; M dmax represents the maximum direct yaw moment intervention value; abs() represents the absolute value function; sgn() represents the sign function.

[0120] Specifically, in order to calculate the direct yaw moment more smoothly and accurately, thereby improving the steering performance and stability of the vehicle, this embodiment uses a high-order fitting curve to solve the intervention degree of the yaw moment, and converts the equivalent required steering angle δ into d Mapping to direct yaw moment M zd middle.

[0121] Step S4: using the required traction torque and the required vehicle speed as indexes, selecting the current optimal driving mode of the unmanned vehicle from a pre-constructed optimal driving mode partition table.

[0122] Specifically, in order to optimize the control process, based on the vehicle's driving mode and motor efficiency distribution data, this embodiment maps different operating conditions to the optimal driving mode through offline analysis and testing, and constructs an optimal driving mode partition table to ensure that the driving mode with the lowest energy consumption and the best performance can be selected under various operating conditions.

[0123] Furthermore, the optimal drive mode partition table is pre-constructed through the following steps S31 to S33:

[0124] Step S31, obtaining various driving modes and different driving conditions of the unmanned vehicle.

[0125] Specifically, all the driving modes of the unmanned vehicle are obtained according to its design and configuration. For example, the four axles of the distributed electric drive 8×8 wheeled vehicle from front to back are 1, 2, 3, and 4, and each axle includes two left and right tires. Its driving modes include four-axle full-wheel drive mode, three-axle (1st axle, 3rd axle, 4th axle) driving mode, two-axle (1st axle and 4th axle) driving mode, and single-axle (4th axle) driving mode.

[0126] The driving conditions include the torque and speed of the unmanned vehicle.

[0127] Step S32: For each driving condition, calculate the energy consumption efficiency of the unmanned vehicle in each driving mode and normalize each driving condition.

[0128] Specifically, under each driving condition, the energy consumption efficiency of the unmanned vehicle in all driving modes is calculated, and the torque and speed of each driving condition are normalized based on the maximum torque and maximum speed in the driving condition to obtain an energy consumption efficiency diagram of each driving condition in different driving modes.

[0129] Step S33: For each driving condition, select the driving mode with the highest energy efficiency as the optimal driving mode, and obtain an optimal driving mode partition table with each driving condition as an index.

[0130] Specifically, Figure 2 As shown, the driving mode with the highest efficiency under each driving condition is statistically normalized, and an optimal driving mode efficiency distribution diagram including each driving mode is drawn.

[0131] More specifically, the optimal driving mode of the current unmanned vehicle is determined based on the required traction torque and required vehicle speed of the unmanned vehicle, and the driving mode is expressed using the following formula:

[0132] S=diag(1,…,λ ij ,…,1),i=1,2,3,4,j=1,2

[0133] Wherein, S represents the driving mode matrix of the unmanned vehicle; i represents the i-th driving axis of the unmanned vehicle; j represents the left wheel and the right wheel of each driving axis of the unmanned vehicle, j=1 represents the left wheel, and j=2 represents the right wheel; λ ij represents the driving state of the jth wheel of the i-th driving shaft. ij =1, it means the jth wheel of the ith drive shaft is the driving wheel. ij When =0, it means that the jth wheel of the i-th driving shaft is a non-driving wheel.

[0134] Exemplarily, when the optimal driving mode of the unmanned vehicle is a three-axis (1st axis, 3rd axis, 4th axis) driving mode, the driving mode matrix of the unmanned vehicle is expressed as S=diag(1,1,0,0,1,1,1,1), and the wheels on both sides of the second driving axle do not participate in the driving.

[0135] Step S5: Under the optimal driving mode, based on the direct yaw moment required by the unmanned vehicle, the output torque distribution result of each driving motor is obtained and distributed to the corresponding driving wheels.

[0136] Furthermore, the optimal driving mode includes a combination of driving motors involved in the driving.

[0137] The step of obtaining the output torque distribution result of each driving motor includes:

[0138] Based on the required traction torque, direct yaw moment and torque output values ​​of each driving motor involved in the driving of the unmanned vehicle, an objective function and corresponding constraints are constructed with the goal of minimizing the weighted sum of torque distribution error and tire load.

[0139] Based on the constraint conditions, the objective function is solved using a quadratic programming algorithm to obtain an output torque distribution result of each driving motor involved in the driving.

[0140] Specifically, the quadratic programming algorithm is a control allocation method with the L2 norm as the main form of the objective function. Quadratic programming takes into account both the equality constraints and inequality constraints of the system, and can handle various physical constraints of the actuator well without the need to specially design an iterative algorithm to handle the actuator constraints and saturation conditions.

[0141] In this embodiment, the controller aims to optimize the combination of torque distribution error and tire load, and reduces the torque distribution error and the total tire load by optimizing the tire torque distribution.

[0142] Furthermore, the objective function is expressed as:

[0143]

[0144] u=S·u′

[0145] v d =[T d M zd ] T

[0146] Wherein, B represents the efficiency matrix of the unmanned vehicle; u represents the output torque distribution vector of each driving motor involved in the driving of the unmanned vehicle; S represents the driving mode matrix of the unmanned vehicle; u′ represents the torque output value of all driving motors of the unmanned vehicle; v drepresents the control output demand expectation matrix; W1 represents the weight matrix of torque distribution error; W2 represents the weight matrix of tire load; T d represents the required traction torque of the unmanned vehicle; M zd represents the direct yaw moment; δ d represents the equivalent required steering angle of the unmanned vehicle.

[0147] Specifically, u represents the output torque distribution vector of each driving motor involved in driving the unmanned vehicle. For example, the torque output value u′ of all driving motors of the unmanned vehicle of the distributed electric drive 8×8 wheeled vehicle is expressed as u′=[T l1 T r1 T l2 T r2 T l3 T r3 T l4 T r4 ] T , where T l1 Indicates the output torque of the left drive motor of the first drive shaft; T r1 Indicates the output torque of the right drive motor of the first drive shaft; T l2 Indicates the output torque of the left drive motor of the second drive shaft; T r2 Indicates the output torque of the right drive motor of the second drive shaft; T l3 Indicates the output torque of the left drive motor of the third drive shaft; T r3 Indicates the output torque of the right drive motor of the third drive shaft; T l4 Indicates the output torque of the left drive motor of the 4th drive shaft; T r4 represents the output torque of the right drive motor of the fourth drive shaft. In the optimal mode S, the output torque distribution vector u of each drive motor involved in the driving of the unmanned vehicle is expressed as u=[T l1 T r1 0 0 T l3 T r3 T l4 T r4 ] T , it can be seen that the motor torque of the second drive shaft is set to 0, indicating that they do not participate in the drive.

[0148] It should be noted that the unmanned vehicle effectiveness matrix B is used to convert the control input u into the motion state of the unmanned vehicle, which is expressed as:

[0149]

[0150] Where d is the distance between the drive shafts; R is the tire radius; δ d represents the equivalent required steering angle.

[0151] Each column of the unmanned vehicle performance matrix B corresponds to the driving state of a tire, and each row represents an aspect of the corresponding vehicle motion state. In this embodiment, the first row of the unmanned vehicle performance matrix B represents the longitudinal speed of the unmanned vehicle, where 1 indicates that the torque of the corresponding tire directly contributes to the longitudinal speed of the vehicle, and the last two values ​​of the first row represent the degree of influence of the steering angles of the left and right steering wheels of the fourth drive shaft on the longitudinal motion of the vehicle; the second row of the unmanned vehicle performance matrix B represents the contribution of the output torque of the unmanned vehicle to the yaw moment of the vehicle; the yaw moment is generated by the torque difference between the left and right wheels, so the value of the left wheel is positive and the value of the right wheel is negative, and the last two values ​​of the first row represent the influence of the steering angles of the left and right steering wheels of the fourth drive shaft on the yaw moment.

[0152] The weight matrix W1 of the torque distribution error is used to adjust the relative importance of different control errors in the objective function. In this embodiment, the weight matrix W1 of the torque distribution error is expressed by the following formula:

[0153] W1=diag(m i ) 8×8

[0154] Among them, m i The weight of the torque distribution error of the i-th drive shaft is determined based on the comprehensive consideration of the path tracking control accuracy and the vertical load rate of the tire. For example, m i Can be set to 0.6.

[0155] The tire load weight matrix W2 is used to adjust the relative importance of tire load distribution in the objective function to ensure that the tire adhesion is reasonably utilized, thereby improving the stability and handling of the vehicle. In this embodiment, the tire load weight matrix W2 is expressed by the following formula:

[0156]

[0157] Among them, n i represents the distribution weight coefficient of the i-th driving axis, which is determined based on the comprehensive consideration of the path tracking control accuracy and the tire vertical load rate. For example, n i Can be set to 0.4; μ i The ground adhesion coefficient of the i-th drive axle is the degree of friction between the tire and the ground, which is obtained based on experience; F zirepresents the tire vertical force of the ith drive shaft, which is the vertical pressure of the tire on the ground. It should be noted that in actual applications, the left and right wheels of the same drive shaft are usually under similar road conditions, so their ground adhesion coefficients and tire vertical forces are similar. In order to simplify calculation and improve control efficiency, this embodiment treats the left and right wheels of the same drive shaft as a group for unified processing when calculating the weight matrix W2 of the tire load.

[0158] Furthermore, when solving the objective function, corresponding constraints need to be formulated, and the constraints include:

[0159] The resultant force of the longitudinal force and the lateral force of each tire involved in driving the unmanned vehicle does not exceed the adhesion limit of the tire.

[0160] Specifically, the tire friction ellipse constraint is based on the adhesion limit of the tire, ensuring that the tire does not exceed its adhesion limit under the combined action of the longitudinal force and the lateral force, ensuring that the tire maintains good adhesion and avoids slipping or loss of control. The tire friction ellipse constraint is expressed by the following formula:

[0161]

[0162] Among them, F xij ,F yij ,F zij They represent the longitudinal force, lateral force and vertical load of the jth wheel of the ith driving axle respectively; C xij ,C yij They represent the lateral stiffness and longitudinal stiffness of the tire of the jth wheel of the i-th drive shaft, respectively, and their values ​​are provided by the tire manufacturer; μ ij Represents the ground adhesion coefficient of the jth wheel of the i-th drive axle.

[0163] The output torque of each wheel hub motor driving the unmanned vehicle does not exceed the adhesion limit of the tire.

[0164] The output torque and torque change rate of each wheel hub motor driven by the unmanned vehicle do not exceed the physical limits of each motor.

[0165] Specifically, the output torque constraint of the hub motor is that the actual output torque of each hub motor must be within the allowable range of the motor's physical properties and vehicle operating conditions to prevent motor overload, motor damage, and tire slippage due to excessive motor output torque, affecting the vehicle's handling and stability.

[0166] More specifically, the following formula is used to express the output torque constraint formula of each wheel hub motor driving the unmanned vehicle:

[0167]

[0168] Among them, T ij Represents the actual output torque of the jth hub motor on the i-th axis.

[0169] The following formula is used to express the output characteristic constraints of each wheel hub motor involved in driving the unmanned vehicle:

[0170] T ijmin ≤T ij ≤T ijmax

[0171] Among them, T ijmin represents the minimum output torque of the jth wheel hub motor of the i-th axis; T ijmax Represents the maximum output torque of the jth hub motor on the i-th axis.

[0172] More specifically, the response characteristic of the wheel hub motor is constrained to ensure that the rate of change of the motor output torque does not exceed a preset value, so as to avoid excessive torque changes that may cause motor overload or vehicle loss of control.

[0173]

[0174] Where, ΔT ij (t) represents the change in motor torque of the jth hub motor of the i-th axis within time Δt; k min Indicates the minimum value of the torque change rate that the hub motor can output; k max It indicates the maximum value of the output torque change rate that the hub motor can output; the maximum and minimum values ​​of the output torque change rate are obtained through experimental tests.

[0175] Optionally, after solving the output torque distribution result of each wheel hub motor, the actual output torque of each wheel hub motor is obtained by collecting the current sensor carried by the unmanned vehicle, and compared with the output torque distribution result to determine whether the wheel hub motor fails, and construct a failure factor matrix. When the wheel hub motor involved in the drive fails, the objective function is solved again based on the failure factor matrix using a quadratic programming algorithm to obtain a new output torque distribution result for each drive motor.

[0176] Specifically, the failure factor of each hub motor is calculated using the following formula:

[0177]

[0178] Among them, η ij represents the failure factor of the jth hub motor of the i-th axis; T i ' j represents the actual output torque of the jth wheel hub motor on the i-th axis; T ij represents the output torque allocated to the jth wheel hub motor of the i-th axis; when η ij=1, it means the jth wheel hub motor of the i-th axis works normally; when η ij When =0, it means that the jth wheel hub motor of the i-th axis has completely failed.

[0179] The failure factor matrix F is expressed as follows:

[0180] F=diag(1,…,η ij ,…,1)

[0181] Based on the failure factor matrix F, the new control effectiveness matrix is ​​calculated using the following formula: F =B·F, so that it replaces the efficiency matrix B in the objective function, adjusts the torque distribution error when a system failure occurs, uses a quadratic programming algorithm to recalculate the output torque of the wheel hub motor involved in the drive, and obtains the distribution result under the failure condition to meet the vehicle driving stability and the vehicle's dynamic performance.

[0182] Another embodiment of the present invention discloses a distributed electric drive unmanned vehicle torque distribution control system with energy consumption optimization, such as Figure 3 As shown, the system includes:

[0183] A tracking error calculation module 101 is used to obtain a speed tracking error and a path tracking error of the unmanned vehicle based on a current driving state and an expected driving state of the unmanned vehicle;

[0184] A lateral / longitudinal tracking control module 102, configured to obtain a required traction torque and an equivalent required steering angle of the unmanned vehicle based on a speed tracking error and a path tracking error obtained by the tracking error calculation module;

[0185] A compound steering equivalent steering angle matching calculation module 103 is used to obtain the direct yaw moment required by the unmanned vehicle based on the equivalent required steering angle obtained by the lateral / longitudinal tracking control module and the lateral stability constraint of the unmanned vehicle;

[0186] A driving mode selection module 104, configured to select the current optimal driving mode of the unmanned vehicle from a pre-constructed optimal driving mode partition table using the required traction torque and required vehicle speed obtained by the lateral / longitudinal tracking control module as indexes;

[0187] The torque distribution module 105 is used to obtain the output torque distribution result of each driving motor based on the direct yaw torque required by the unmanned vehicle in the optimal driving mode and distribute it to the corresponding driving wheels.

[0188] The above system embodiment and method embodiment are based on the same principle, and their related parts can be used for reference, and can achieve the same technical effect. The specific implementation process refers to the above embodiment, which will not be repeated here.

[0189] In summary, the energy-optimized distributed electric drive unmanned vehicle torque distribution control method and system according to the embodiment of the present invention has the following beneficial effects:

[0190] 1. The present invention realizes the integrated control of path tracking and energy management through a hierarchical control framework. The upper-level controller is responsible for determining the required total driving torque and yaw moment, the middle-level controller controls the torque distribution between axles according to the energy consumption, and the lower-level controller refines the torque distribution of each driving wheel based on vehicle stability and path tracking performance, thereby realizing effective decoupling of energy management and torque distribution, which can not only improve the path tracking performance of the unmanned platform, but also realize efficient energy management.

[0191] 2. The present invention calculates the energy consumption efficiency distribution diagram under different driving modes offline through bench testing, and then determines the optimal motor operating area through the speed and required torque index, ensuring that the motor operates under high-efficiency conditions and can ensure that the torque is distributed to maximize the energy conversion efficiency.

[0192] 3. The present invention takes into account factors such as tire vertical load distribution and path tracking error, and adopts a quadratic programming algorithm to achieve output torque distribution of each drive motor as well as steering angle and braking torque, which can effectively improve the driving control stability of the unmanned platform off-road maneuverability.

[0193] 4. The present invention introduces a fault-tolerant control mechanism, which monitors the fault status of each drive motor in real time and recalculates the control allocation according to the failure factor matrix to ensure that the stability and reliability of the vehicle can be maintained in the event of a motor failure, and has higher adaptability and robustness.

[0194] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A torque distribution control method for a distributed electric drive unmanned vehicle with energy consumption optimization, characterized in that: The steps include: Get the current driving status of the unmanned vehicle; Based on the current driving state and the expected driving state of the unmanned vehicle, obtaining the required traction torque and the equivalent required steering angle of the unmanned vehicle; Based on the equivalent required steering angle and the lateral stability constraint of the unmanned vehicle, obtaining a direct yaw moment required by the unmanned vehicle; Using the required traction torque and the required vehicle speed as indexes, selecting the current optimal driving mode of the unmanned vehicle from a pre-constructed optimal driving mode partition table; In the optimal driving mode, based on the direct yaw torque required by the unmanned vehicle, the output torque distribution result of each driving motor is obtained and distributed to the corresponding driving wheels.

2. The method according to claim 1, characterized in that: The driving state includes a driving speed and a driving speed azimuth; The step of obtaining the required traction torque and the equivalent required steering angle of the unmanned vehicle includes: Based on the current driving speed and the expected driving speed of the unmanned vehicle, a speed tracking error of the unmanned vehicle is obtained; Based on the speed tracking error, obtaining a required traction torque of the unmanned vehicle; Obtaining a path tracking error of the unmanned vehicle based on the current driving speed azimuth of the unmanned vehicle and the tangent azimuth of the road at the preview point; Based on the path tracking error, an equivalent required steering angle of the unmanned vehicle is obtained.

3. The method according to claim 2, characterized in that: The obtaining of the path tracking error of the unmanned vehicle includes: Based on the current driving speed azimuth of the unmanned vehicle and the tangent azimuth of the road at the preview point, an equivalent reference fusion deviation angle of the unmanned vehicle is obtained; Based on the current position of the unmanned vehicle and the position of the preview point, obtaining a lateral position deviation of the unmanned vehicle; Based on the equivalent reference fusion deviation angle and the lateral position deviation, obtaining a calculated value of an equivalent fusion deviation angle of the unmanned vehicle; Based on the equivalent reference fusion deviation angle and the calculated equivalent fusion deviation angle, a path tracking error of the unmanned vehicle is obtained.

4. The method according to any one of claims 2 or 3, characterized in that: Based on the speed tracking error, using PI control to obtain the required traction torque of the unmanned vehicle; Based on the path tracking error, PI control is used to obtain the equivalent required steering angle of the unmanned vehicle.

5. The method according to claim 1, characterized in that: The obtaining of the direct yaw moment required by the unmanned vehicle based on the equivalent required steering angle and the lateral stability constraint of the unmanned vehicle comprises: Based on the lateral stability constraint of the unmanned vehicle, a maximum direct yaw moment intervention value is obtained; The intervention degree of the yaw moment is solved using a high-order fitting curve, and the direct yaw moment required by the unmanned vehicle is obtained using the following formula: M zd =sgn(δ d )(-2D 3 +3D 2 )M dmax Δ=abs(δ d / d max ) Among them, M zd represents the direct yaw moment; δ d represents the equivalent required steering angle; δ max represents the maximum mechanical steering angle of the unmanned vehicle; M dmax represents the maximum direct yaw moment intervention value; abs() represents the absolute value function; sgn() represents the sign function.

6. The method according to claim 1, characterized in that: The optimal driving mode includes a combination of driving motors involved in driving; The step of obtaining the output torque distribution result of each driving motor includes: Based on the required traction torque, direct yaw moment and torque output values ​​of each driving motor involved in the driving of the unmanned vehicle, an objective function and corresponding constraint conditions are constructed with the goal of minimizing the weighted sum of torque distribution error and tire load; Based on the constraint conditions, the objective function is solved using a quadratic programming algorithm to obtain an output torque distribution result of each driving motor involved in the driving.

7. The method according to claim 6, characterized in that: The objective function is expressed as: u=S·u′ v d =[T d M zd ] T Wherein, B represents the efficiency matrix of the unmanned vehicle; u represents the output torque distribution vector of each driving motor involved in the driving of the unmanned vehicle; S represents the driving mode matrix of the unmanned vehicle; u′ represents the torque output value of all driving motors of the unmanned vehicle; v d represents the control output demand expectation matrix; W1 represents the weight matrix of torque distribution error; W2 represents the weight matrix of tire load; T d represents the required traction torque of the unmanned vehicle; M zd represents the direct yaw moment; δ d represents the equivalent required steering angle of the unmanned vehicle.

8. The method according to claim 7, characterized in that: The constraints are: The resultant force of the longitudinal force and the lateral force of each tire involved in driving the unmanned vehicle does not exceed the adhesion limit of the tire; The output torque of each wheel hub motor driving the unmanned vehicle does not exceed the adhesion limit of the tire; The output torque and torque change rate of each wheel hub motor driven by the unmanned vehicle do not exceed the physical limits of each motor.

9. The method according to claim 1, characterized in that: The optimal drive mode partition table is pre-built by the following steps: Obtaining various driving modes and different driving conditions of the unmanned vehicle; For each driving condition, calculating the energy consumption efficiency of the unmanned vehicle in each driving mode and normalizing each driving condition; For each driving condition, the driving mode with the highest energy efficiency is selected as the optimal driving mode, and an optimal driving mode partition table with each driving condition as an index is obtained.

10. A distributed electric drive unmanned vehicle torque distribution control system with energy consumption optimization, characterized in that: include: A tracking error calculation module, used to obtain a speed tracking error and a path tracking error of the unmanned vehicle based on a current driving state and an expected driving state of the unmanned vehicle; A lateral / longitudinal tracking control module, configured to obtain a required traction torque and an equivalent required steering angle of the unmanned vehicle based on a speed tracking error and a path tracking error obtained by the tracking error calculation module; A compound steering equivalent steering angle matching calculation module, used to obtain the direct yaw moment required by the unmanned vehicle based on the equivalent required steering angle obtained by the lateral / longitudinal tracking control module and the lateral stability constraint of the unmanned vehicle; A driving mode selection module, configured to select the current optimal driving mode of the unmanned vehicle from a pre-constructed optimal driving mode partition table using the required traction torque and required vehicle speed obtained by the lateral / longitudinal tracking control module as indexes; The torque distribution module is used to obtain the output torque distribution result of each drive motor based on the direct yaw torque required by the unmanned vehicle in the optimal driving mode and distribute it to the corresponding drive wheels.

Images