Development On Control System Of Chassis Domain For Distributed Drive Intelligent Vehicles

In recent years,intelligent vehicle has gradually become a hot research topic.In the existing research on intelligent vehicle trajectory tracking control,most of the research focuses on improving the performance of trajectory tracking,while ignoring the vehicle handling and stability in the tracking process under extreme conditions.Distributed drive electric vehicle is driven straightly by in-wheel motors,which has the characteristics of high controllability,rapid response and good chassis dynamics expansion performance.Research of distributed drive electric vehicle on the coordinated control of trajectory tracking and stability can provide safe and reliable dynamic response for high-level intelligent driving,and realize reasonable division of labor with intelligent driving domain.Aiming at the above problems,this paper will design a chassis domain control system for distributed drive intelligent vehicle.The specific research in this paper is as follows:In order to facilitate the development of control strategy and simulation verification,a7 DOF vehicle dynamics model based on Uni-Tire tire model and a joint Carsim and Matlab/Simulink simulation model are established,and the accuracy of the two models is verified by using the data of angular pulse working condition of real vehicle.Then,considering the problem that it is difficult to obtain the actual vehicle speed and road adhesion coefficient,based on the 7DOF vehicle model,a nonlinear vehicle speed observer based on tire force is designed to estimate the longitudinal and lateral vehicle speed.In addition,the Extended Kalman Filter is used to estimate the road adhesion coefficient.Then it discusses and summarizes the relationship between vehicle instability boundary with road adhesion coefficient,vehicle speed and front wheel angle.It lays a foundation for the subsequent development of chassis domain system.Then a layered chassis domain control system is proposed.The upper layer carries out trajectory following control,the middle layer carries out yaw stability control,and the bottom layer is used to coordinate the middle and upper layer as well as distribute torque.The upper trajectory following control module decouples the longitudinal and lateral control,and sliding mode control algorithm based on feedback linearization is used to follow the desired speed longitudinally.Laterally,adaptive pre-aiming feed-forward and PID feedback control algorithm is used for path point following.The middle layer provides stability control,an integrated controller combining active front wheel steering control and torque vector control is designed to calculate the additional yaw moment and the additional front wheel steering angle which is required to maintain vehicle stability,and the integrated controller uses model prediction control algorithm.The bottom torque distribution module performs coordinated control.Based on the current vehicle condition and road condition,an optimized torque distribution algorithm is designed according to the road attachment ellipse,which considers both the upper trajectory follow-up and the middle trajectory stability.The control weight of upper trajectory following and middle trajectory stability control are adjusted adaptively under extreme conditions.Based on the weight,the control quantity determined by the upper and middle level decision is weighted,and then the weighted control quantity is allocated to the chassis actuator by multi-objective optimization to obtain the control quantity value finally sent to the chassis actuator.Finally,the control algorithm is simulated and verified based on the established joint simulation platform.The simulation result shows that the chassis domain control system can improve the handling stability of the vehicle whether on high attached or on low attached roads,and also play the role of collaborative intelligent driving domain and chassis dynamic domain.The hardware in the loop platform is built,and the control system is deployed to VCU in the form of embedded code for online verification.The experimental result indicates that the experimental is similar to the simulation,which verifies the effectiveness and engineering implementation ability of the algorithm.