Research On Motion Control Method Of Four Wheel Independent Drive Vehicle In Complex Ground Environment

With the continuous development of intelligent control technology,the application fields and application environment of wheeled intelligent vehicles are constantly expanding.These new application fields or environments put forward higher requirements for the flexibility and mobility of vehicles.The traditional wheeled vehicles can not meet these requirements because of their own shortcomings.The multi-wheel independent drive and independent steering vehicle based on the integrated electric wheel has unique advantages in terms of motion flexibility,mobility and environmental adaptability.Therefore,the research on the integrated electric wheel and its multi-wheel integration and control technology has become one of the research hotspots in the field of intelligent electric drive vehiclesThis paper focuses on the key problem of how to perceive and accurately understand the ground environment information when the four wheel independent drive vehicle based on the integrated electric wheel performs the operation task in the complex ground environment,so as to complete the operation task autonomously,efficiently and stably.This paper studies the motion control method of four wheel independent drive vehicle in complex ground environment and the research includes the following aspects:1.In order to realize the vehicle moving stably and efficiently in complex ground environment,a method of vehicle speed adaptive adjustment based on vehicle body vibration information is proposed.In this method,the GMM model of typical ground is constructed by using the vibration information of vehicle body,and the model parameters are solved based on the improved EM algorithm,and the clustering analysis of arbitrary terrain is realized efficiently and quickly.The analysis results are combined with the fuzzy control based on terrain slope fluctuation to realize the adaptive control of vehicle motion speed and improve the vehicle motion efficiency and ground adaptability2.The method of using image information to pre-judge the ground category and carry out motion planning is proposed,which solves the problem that only using the vibration information can not accurately judge the terrain change trend,which leads to the delay of speed adjustment.Using CNN model,based on multi task learning method,the type of ground image is identified and the relative distance of ground switching is estimated.Before the vehicle enters the new ground type,the speed is planned by using the quintic polynomial motion planning method.The smooth transition of speed under different ground types is realized,and the stability of vehicle motion is further improved3.The control effect of vehicle suspension is affected by many factors,such as suspension model,vehicle body state,ground type and so on.Considering that the control of suspension system is a continuous decision-making process,this paper proposes a semi-active suspension control strategy based on improved DDPG deep reinforcement learning algorithm,which makes full use of the continuous learning ability of the algorithm to automatically adjust the suspension control parameters and improve the suspension control.Finally,the simulation results verify the feasibility and superiority of this method4.In view of the time-varying characteristics of the vehicle steering process and the error interference in the establishment of steering model,a robust sliding mode controller with strong disturbance suppression ability and strong input tracking ability is designed for vehicle steering given tracking control.The simulation results show that this method is suitable for the vehicle steering control5.In order to improve the braking performance of vehicles,a braking control method based on real-time estimation of road adhesion coefficient is proposed.The disturbance observer with compensation is designed to estimate the road adhesion coefficient,which provides the basis for the given slip rate in the braking process.Then,the improved FOA algorithm is used to optimize the parameters of PID brake controller.The simulation results show that the PID controller is better than the classical Bang-Bang controller,the PID controller based on genetic algorithm and the PID controller based on Z-N method.