Study On Kinematic Intelligent Control Method For Outdoor Wheeled Mobile Robot

Wheeled mobile robot is an integrated system, which has many kinds of functions,such as external environment perception, path planning, navigation, decision motioncontrol and implementation. Meanwhile, due to the simple structure and flexiblemovement, wheeled mobile robot is widely used in many fields. However, when itcomes to outdoor complex environment, the robot is inevitably affected with uncertainexternal disturbances, load changes and the dead-zone of interference, which may havean effect on the stability of the system. Therefore, how to suppress these disturbances isthe key to improve control performance of the outdoor wheeled mobile robot motion.This dissertation focuses on the kinematic characteristics of wheeled mobile robots,and then proposes adaptive fuzzy dead-zone compensation by establishing thekinematic model based on the path deviation. Moreover, according to parameterperturbation and external disturbance which appear in the movement of the wheeledmobile robots, the corresponding control strategy is put forward combining multi-modecontrol and fuzzy-PID control to improve the dynamic properties of kinematic control.The main works are as follows:Firstly, the kinematic model and dead-zone characteristics analysis of robot systemare studied as a function of the deviation between motion state and the displacement andangle,Secondly, based on fuzzy control and on-line prediction, an adaptive dead-zonecompensation of the control system is proposed taking effects of unknown dead-zoneinto consideration. Consequently, the Lyapunov stability is conducted to validate theconvergence of the tracking error.Then, since the parameter perturbation and external disturbance will appear in therobot movement, the corresponding control algorithm is put forward combiningmulti-mode control and fuzzy-PID control. Specifically, considering the characteristicsof feature identification，an appropriate control strategy is selected, which can tuningcontrol parameters on-line based on fuzzy rules.Lastly, the simulation of comparison with traditional PID control and fuzzy PID iscarried out and the results show that the proposed method can compensate the effects ofparameter perturbation and improves the control performance of the robot motioneffectively.