Research On Motion Control Algorithm Of The Two-wheeled Robot

Robot has been playing an important role since it was invented,especially widely used in manufacturing,production,military,production and life,etc.Therefore,the research of robot has both theoretical value and commercial prospect.Practical application value.The two-wheeled robot is a special member of the robot.Due to its flexibility and convenience,it has been widely studied and studied by domestic and foreign researchers.This special robot is extremely suitable for dangerous goods transportation,short-distance mobility tools,etc.However,due to the existence of two-wheeled robot itself The characteristics of nonlinearity,instability,and strong coupling make the design of the controller more difficult.Therefore,the study of two-wheeled robotic systems has both theoretical and practical implications.In this paper,the rigid two-wheeled robots are studied,that is,the left and right wheels and the car body are independent.Firstly,the force of the two-wheeled robots is analyzed,and then the kinematic model and dynamics model are established using Newtonian mechanics method and Lagrangian method respectively.The two-wheeled robot model has nonlinearity and strong coupling.Therefore,in order to facilitate the analysis of the two-wheeled robot system,the model needs to be linearized and decoupled.Finally,the stability of the two-wheeled robot model is analyzed and determined by control theory.The controllability and observability provide a theoretical basis for the design of the controller below.The attitude information of the two-wheeled robot is collected by combining the gyroscope and the accelerometer.In order to achieve the accuracy of information collection and the control accuracy of the robot,the Kalman filter algorithm is used to fuse the data collected by the two sensors to obtain two-wheeled robots.Precise gesture information.Based on accelerometer and gyroscope models,use SIMULINK simulation analysis.The simulation results verify the filtering effect of the Kalman filter.Based on the linear model of the two-wheeled robot,the pole configuration theory and the LQR optimal controller theory were used to analyze and design the corresponding controller.Comparing the performance of the two controllers using SIMULINK,the results showed that the LQR controller can restore the upright balance of the two wheeled-robot;however,the weight matrix Q and R values are selectedbased on human experience when designing the LQR controller,it is subjective,which make the performance of the designing control not reach the best state.Based on this defect,this paper uses the particle swarm optimization algorithm to globally optimize the weight matrix,obtains an optimal set of weight matrix,solves the optimal feedback control rate K,and finally uses the SIMULINK to optimize the LQR.The performance of the controller is verified.