Research On Trajectory Tracking Control Of Wheeled Mobile Robot Under Sliding

With the development of computer science and artificial intelligence technology,the research on robots has been attached more and more attention.Compared with traditional industrial robots,wheeled mobile robots are generally used in complex environment to accomplish tasks.In an unknown environment,skidding and slipping phenomena are inevitable for them so it is of major significance to study the trajectory tracking control of wheeled mobile robots in slipping.This paper mainly takes the nonholonomic wheeled mobile robots with wheel skidding and slipping as the research objects,and combines with the method of backstepping control,robust adaptive control,Gaussian wavelet network,adaptive fuzzy control,disturbance observer and Lyapunov stability theory to study their trajectory tracking control.The main content of this paper includes the following four aspects:(1)For the kinematics model of mobile robot under sliding,firstly,considering the influence of side slip,the kinematic controller is designed by Lyapunov and backstepping control.Then considering the existence of both wheel slip and side slip,the robust adaptive controller is designed by robust adaptive method,and the disturbance effect of the moving robot is compensated by the robust adaptive parameters.Finally,simulation experiments show that the robust adaptive controller can better compensate the system for the effects of slip and side slip.(2)For the kinematics model of a wheeled mobile robot in slipping,firstly,auxiliary kinematics controller is designed to deal with the errors between the wheeled mobile robot's attained pose and expected pose caused by slipping and skidding,the wheeled mobile robot's speed is obtained.Then dynamic controller is designed with the method of Gaussian wavelet network and adaptive learning method among which the former is used to approximate the system's unknown non-linearity while the later to update the dilation and translation parameters of Gaussian wavelet function online.At last,considering the large amount of computing of Gaussian wavelet network based controller designed,similarly,this chapter also uses the adaptive fuzzy method with the same approximation ability in accuracy to design an adaptive fuzzy dynamic controller.An entire control system is obtained by combing the auxiliary kinematic controller and adaptive fuzzy controller.Through comparison with the traditional method of adaptive control,it is displayed that the controller designed can help quickly and accurately track the trajectory of the robot in slipping.(3)Disturbance observers are introduced in the tracking control of wheeled mobile robots' trajectory in slipping.A nonlinear disturbance observer and fuzzy disturbance observer are designed respectively.Through the online estimation of the nonlinear disturbance observer on skidding disturbance and compensation to the kinematic controller,the speed of the wheeled mobile robot is obtained.The disturbance observer is designed based on the rapid approximation ability of the fuzzy method to observe the lumped disturbance and compensate it to the dynamic controller,and the wheeled mobile robot's control moment is obtained and trajectory tracking achieved in slipping.At last,system simulation is carried out of the two disturbance observers to verify the control effectiveness.(4)The adaptive fuzzy dynamic controller and fuzzy disturbance observer designed previously are respectively validated on the experimental platform of wheeled mobile robots.Finally,according to all the experimental results,the feasibility and effectiveness of the controllers designed above mentioned are validated.