Motion Control Research Of Two-Wheel Robot Based On Human-Simulated Intelligent Control

Mobile robot is a comprehensive system consisting of many functions such asenvironment awareness, dynamic decision and planning,motion control and taskexecution. With constant improvement in performance,mobile robot,particularly theservice-oriented mobile robot, will enjoy wide market prospects. Control problemsconcerning two-wheeled robot mainly contain three aspects including pointstabilization, path planning and path tracking. Point stabilization underlying thesolution to the basic motion problems,path planning and path tracking.The two-wheeled differential drive mobile robot is subject to nonholonomicconstraints.Besides, the motor drive circuit of the system contains a doubleclosed-loop control structure with saturation nonlinear links.Thus the pointstabilization control of mobile robot can be transformed to studies on control problemsof a multiinput—multioutput （MIMO）nonlinear system, and has been a hot area ofresearch for several years.As a system subject to classical nonholonomic constraints,it can not satisfy therequirement of Brockett smooth continuity stabilization, i.e. no continuitytime-invariant state feedback control law as a method to achieve point stabilizationcontrol of the two-wheeled robot.Meanwhile, the saturation linear links in two motorsystem greatly make system control more difficult.Aiming at point stabilization control of two-wheeled mobile robot, this papermainly finishes the following works:①The DC motor double closed loop drive control method with adjustabledynamic parameters is proposed on the basis of human-simulated intelligent control.The author has designed and made the new type of DC motor double-closed loopdriver and then complete the effective control of upper limit current of motor drivesystem.②ntelligent point stabilization controller with three-layer structure is proposedfor the two-wheeled differential drive mobile robot.The top layer of the controlleremploy a multi-modal human-simulated controller based on anger and distance errorcharacteristic to effectively solve the nonholonomic constraint problem ofkinamatics.The middle layer adopt a multi-modal current limitation controller toachieve a effect coordination of motor system,so as to overcome the incontrollable behavior caused by saturation nonlinear links.The bottom layer use a drive system inwhich the dynamic parameters can be adjusted to accomplish the drive of the robot.The drive system has a configuration of double-closed loop controller.The speedloop controller is designed with the human-simulated intelligent method, possessingthe capacity to choose different control modal in light of speed error. Hall sensors getarmature current as a feedback information to form a current loop controller in whichthe feedback factor can be altered to adjust the upper limit current value of motor drivesystem,so as to obtain adjustable dynamic parameters of the motor.Experiment has proved the feasibility of dynamically adjusting the parametersduring acceleration process of motor and also achieve point stabilization control ofrobot. The three-layer structure point stabilization control method based onhuman-simulated intelligent control is also proved to be effective fortwo-wheeled differential drive mobile robot and DC double-closed loop driver withadjustable dynamic parameters is verified to be practical.