Several Types Of Control Studies On Non-holonomic And Under-actuated Systems

Nonlinear,time-delay and external stochastic disturbance are commonly encountered in real systems and the traditional approaches based on linear theory does not work well on nonlinear systems.Therefore,there still exist some open problems needed to be further investigated in the nonlinear system fields.In this paper,several classes of nonlinear systems with nonholonomic,time-delay,stochastic,underactuated and external disturbance properties have been investigated.The main contributions are outlined as follows:1.For a class of uncertain flexible joint robot system(FJR),a modified adaptive dynamic surface control technique is introduced in this part.By using the adaptive backstepping method,a new adaptive controller is constructed to avoid the overparametrization and explosion of complexity problem.All the signals of the closed-loop system are rendered globally/semi-globally uniformly ultimately bounded.Finally,a simulation is given to show the validity of the control algorithm.2.For a class of nonholonomic systems with uncertain parameters and delays,the state feedback stabilization problem is studied.For the first time,the adaptive control problem of uncertain and nonholonomic systems with multiple time delays is considered.By introducing a dynamic gain based Lyapunov-Krasovskii method and technique,the corresponding state feedback adaptive controller is designed and constructed,and the designed controller can make the closed-loop system asymptotically stable.3.For a class of nonlinear systems with unknown control coefficient and external disturbance,using specifying the control input method by adding a power integrator to extend the globally finite-time tracking results in the previous studies.The simulation shows that the designed control law guarantees the reference signal can be followed in a fixed time.4.For a class of nonholonomic underactuated cranes with highly,nonlinearly coupled and external disturbances,a combined RBF neural network and adaptive controller is introduced to state feedback control.The use of combined approach compensates for the problems caused by inaccurate model.The actual operating results on the hardware device(RE3DCrane)are provided to verify the practical performance of the control strategies.Finally,a summary has been done in the dissertation and the future work is presented.