Game-based Four-wheel Independent Driving Vehicle Fault-tolerant And Longitudinal & Lateral Motion Integrated Control

The four-wheel independent driving autonomous driving electric vehicle is widely concerned because its four driving wheels can be controlled independently,which greatly improves the flexibility of system control and has considerable development potential.However,due to its more actuators,and the use of x-by-wire technology,increased the possibility of failure.If no effective safety measures are taken,a traffic accident may occur and even the passengers’ life may be threatened when the driving system suddenly fails during driving.Therefore,this paper mainly focuses on fault tolerance and integrated control of longitudinal and lateral motion of four-wheel independent driving electric vehicles after the failure of the driving system.This paper uses cooperative game theory to deal with the cooperative relationship between different control variables in the control process.Based on a brief introduction to the theory of Pareto frontier,threatpoint and bargaining solutions in the cooperative game framework,aiming at the problem of disharmony of the Egalitarian bargaining solution result when the cost gap between players is too large,a solution for cost normalization method is proposed,and the cooperative game theory is applied to four-wheel independent drive electric vehicle driving fault-tolerant and longitudinal and lateral motion integrated control.First,the basic strategy of fault-tolerant integrated motion control based on cooperative game is proposed,that is,the upper-level path tracking controller based on cooperative game generates the front wheel steering angle and additional yaw moment;the upper-level speed tracking controller based on feedforward-feedback generates the desired driving force;the front wheel steering angle generated by the upper controller directly controls the front wheel steering,and the additional yaw moment and the desired driving force are input into the lower layer fault-tolerant controller,the fault-tolerant controller uses the quadratic programming algorithm to redistribute the driving torque under the premise of meeting the control amount input by the upper controller,so as to achieve the purpose of comprehensive control of fault-tolerant and motion.Then,the control strategy is optimized and improved.For the problem that the cooperative game cannot dynamically adjust the cooperative weight in real time,the model predictive control is combined with the cooperative game,and the characteristics of the model prediction control rolling prediction are used to realize the real-time dynamic adjustment of the cooperative game weight;Aiming at the problem that the control output of the feedforward-feedback speed tracking control is not smooth,a speed tracking controller with higher control performance is designed using model prediction control;the fault-tolerant controller based on quadratic programming is changed to the one based on minimax,and the maximum utilization rate of the four driving wheels is reduced;through the above improvements,the performance of the overall control strategy has been significantly improved.Finally,the overall control structure is simplified,and the independent path tracking and speed tracking controllers are improved into longitudinal and lateral coupling motion controllers based on model prediction cooperative games;by changing the control amount into a control increment,the original fault-tolerant controller has the ability to smooth the violently change of the upper control signal and filter the pulse signal;in the end,not only the control structure is simplified,but also the overall control performance is significantly improved,that is,It is to improve the speed and path tracking accuracy while reducing the amount of control.