| The rapid and stable control of power equipment has always been a major research field of robot research,and the rapid development of artificial intelligence is to promote the intelligent algorithm in intelligent control applications.Fuzzy control,neural network adaptive control,genetic algorithm and sliding mode variable structure control technology become more and more mature,and they become solutions to the problem of nonlinear complex system control.The seamlessly link the intelligent algorithm to the wheeled probe robot drive control system to improve its accuracy is still a major research hotspot in the field of control.In this thesis,Solidworks software is used to design the chassis of the wheeled probe robot,and a physical model of the wheeled driving robot is established.According to the mechanical principle of dynamic analysis,the kinematic simulation determines the robot movement characteristics.The appropriate motor model is selected according to the robot’s performance and load requirements.Then a brushless DC motor is selected to drive the robot chassis wheels.The brushless DC motor speed loop controller adopts Anti-windup PI and integral sliding mode variable structure control in two ways.The current loop adopts hysteresis control.By contrast,the integral sliding mode controller is more stable and fast,but there is a chattering phenomenon when the integral sliding mode controller controls the speed loop.In order to solve the phenomenon of chattering,the introduction of a BP neural network forms a composite control.The design of the BP neural network adopts the online updating learning way to adjust the sliding mode variable structure system.The method not only reduces the brushless DC motor control error,but also weakens the brushless DC motor control system chattering problem.In this thesis,a sliding mode neural network controller is used to control the speed loop of the brushless DC motor on the wheeled probe robot.Compared with the simulation results of the integral sliding mode variable structure control,the chattering is improved,and the control precision is improved.The correctness and validity of the design are verified,and the superiority of the control method is proved. |