Optimal Torque Control For Steering Of Multi-axle Vehicle With In-wheel Motors

Compared with the traditional vehicle,due to the configuration characteristics of in-wheel motor drive vehicle,the motor directly drives the wheel,which has many advantages such as low pollution and energy consumption,diverse driving forms,flexible space layout,etc.,it’s widely considered as the final form of vehicle development.It also conforms to the vehicle development trend for the current background such as climate warming,energy crisis and so on.what’s more,it fits the features and requirements of multi-axle heavy vehicles especially military multi-axle vehicles.Multi-axle vehicles play a decisive role in daily life and national construction,but because of its structure,whose kinematics and dynamics characteristics are more complex than two-axle vehicles.Meanwhile,the self-weight and the load of multi-axle vehicle are large,so the demand for energy is more urgent.And battery as the power source of electric vehicles,has long been a bottleneck restricting the development of electric vehicles,so at this stage the battery technology can’t improve substantially,the study of energy-saving control for multi-axle vehicle with in-wheel motors is necessary,which also prepares for the rapid development of the electric vehicle after the battery technology is solved.First of all,the theory of vehicle dynamics is used to divide the whole vehicle into several subsystems and then complete the four-axle vehicle with in-wheel motors dynamics model by Simulink.To verify the correctness and accuracy of the model,the Adams vehicle model with the same parameters is completed.Simulation results show that under various working conditions,the related driving parameters are highly consistent and the maximum error of simulation accuracy is less than 2%,which proves the correctness of the vehicle dynamics model and provides a reliable basis for the simulation of the following text.In order to determine the principle of optimal torque distribution on the energy consumption of the vehicle steering,2 DOF vehicle model is adopted to establish the wheel motion equations.While vehicle turning,changing inside and outside wheel drive torque ratio can produce a yawing moment and it change turning resistance.However,it is not the case that the larger the driving torque of the outer wheel is,the more obvious the energy-saving effect will be.There is an optimal distribution point under different speeds and wheel rotation angles.At the same time,the distribution of torque also affects the working efficiency of the motor,which should be a process of dynamic optimization.The research object is the 8×8 in-wheel motor vehicle,which is far different from the two-dimensional optimization problem of traditional vehicles and should operate an eight-dimensional independent vector.Combined with artificial intelligence,DDPG algorithm is used,the state parameter selects wheel speed and steering angle,the action parameter selects drive torque distribution ratio and the reward parameter selects vehicle SOC,which ensures the maximum reward parameter is the realization principle.The Actor and Critic network are finished in MATLAB,combined with the dynamic model.In the process of training,the parameters of two networks are trained alternately and finally the Actor network embedded into the vehicle model.After training,simulation results show that the maximum energy consumption of the vehicle can be reduced by about 5% in the steering process with the motor efficiency MAP used in this paper.In addition,the vehicle stability problem in the process of turning is also considered.The stability control is carried out by taking the slip angle and yaw velocity into consideration and the desired value is calculated by 2 DOF model.The desired additional yaw moment of the vehicle is obtained by fuzzy PID control.Besides,the two-wheel and multi-wheel coordinated control mode are designed to ensure the stability of the vehicle.In order to avoid the interference of control,the coordinated control strategy is designed to realize the torque optimization control based on energy consumption on the basis of ensuring the stability of the vehicle.On the basis of simulation verification of energy saving and stability control respectively,the whole system including coordination control is simulated and verified,and the test bed is built.The TruckSim vehicle model was downloaded into a Simulator and the control model was downloaded into TTC.By offline simulation and HIL test can prove that the energy consumption of the vehicle control can make the vehicle of SOC reduce 5% at most and don’t intervene with the vehicle stability control under the coordination control,which ensure the vehicle driving stability,so as to achieve torque optimization control purposes for multi-axle vehicle with in-wheel motor.