Research On Design Of Nonholonomic Mobile Robot System Structure And Its Trajectory Tracking Control

In recent years, robotics have entered into the new moment along with subjects such as computer engineering, electronic engineering, control theory, artificial intelligence and sensor theory, etc which have developed and maturated incessantly from programmable playback industry robot to providing with some sensors and adaptive ability, and to intelligent robot equipped multi-advanced sensors, the research tasks have went through a process which is from simple to complex, form function singularity to multiplicity, from industry manufacture to military, nucleus area, aviation and spaceflight, services, medical apparatus, gene project and many areas. Forecast, to soon future, robot applications areas will expand widely and thoroughly. So the applications of robot also bring the new requires to the robotics. And the control of robot come into being under this background, and become the important embranchments of robotics. In this paper, the history, actuality and the classify of mobile robot are reviewed, present the effective methods of trajectory tracking.In this paper, firstly, mobile robot hardware and software structure are introduced, and present the robot how to organize and control the structure finishing the tasks. Secondly, useful nonholonomic wheel mobile robot definition is presented, and analysis kinematic model and dynamic model of mobile robot, then transfer the kinematic model to standard chain system and power equation. Thirdly, according to the kinematic and dynamic model of robot, we design trajectory tracking method based on calculation torque, introduce the control structure and analysis the control method astringency, then simulation result indicate the method effective. But the method can not solve system uncertainties which do great effects on its perfermence. So utilizing the fuzzy controller which does not depend on the accurate mathematics overcomes the effects of the nonlinears, coupling and uncertain factors. Finally, for countertainty the defects of fuzzy control, a fuzzy neural network is designed in this paper, whose network is recursion neural network. The neural network is used to compensate the fuzzy control deficiency, also the constringency is proved, and the stability of the system is guaranteed.All proposed methods can be applied on the robotic trajectory tracking control to counteract the effects of the nonlinear, coupling and uncertain factors in the robotic system in the paper. Combining with the national wide university students smartcar race, we design the hardware and software of the intelligent smartcar, and achieve control method that is proposed have been naturalized in the smartcar. The result of contest indicates control arithmetic effective. Meanwhile, the experiment smartcar provide us a good research platform.