Composite Control And Application Of Semi-Markovian Jump System Based On High-Order Disturbance Observer

With the continuous improvement of the level of automation and intelligence,the high-precision control and high-performance indicators of robots have always been research hotspots in the industrial and academic fields.However,during the actual operation of the system,the internal parameters or structures are easily affected which resulting in random jumps in the system mode.In the engineering field,modeling and control of such systems is very challenging.Semi-Markov jump systems(S-MJSs)can be used to describe such complex systems and thus have received extensive attention in academia and industry.Considering that in the high-precision electronic assembly process,the effective load mass and moment of inertia of the manipulator may undergo sudden changes in parameters,and the controller also needs to be changed accordingly.Therefore,it is advantageous to study this scenario based on the S-MJSs model.In addition,internal and external disturbances are common in robot systems,which often seriously affect the control performance.Therefore,how to design an anti-disturbance control strategy to make the system stable and have high performance becomes particularly important.Based on this,this paper focuses on the composite anti-disturbance control problem of a class of robotic systems under the influence of uncertain harmonic disturbance,norm-bounded disturbance,modal jumps,unknown nonlinear and event-triggered.The main work is as follows:(1)The anti-disturbance control problem of S-MJSs with unknown sinusoidal disturbance is studied.The disturbance here is a modellable matching disturbance with uncertain parameters,which widely exists inside the motors of the robotic arm drive joints.In order to estimate and cancel such disturbance,a novel high-order disturbance observer is constructed,which transforms the estimation problem of the uncertain sinusoidal disturbance into an estimation problem of the uncertain parameter.Aiming at the difficulty of the time-varying transition probability in S-MJSs,the method that the residence time of a given transition probability has upper and lower bounds is used to solve it.Finally,the effectiveness of the proposed control strategy is verified by an example simulation of a one-degree-of-freedom(1-DOF)manipulator.(2)The composite anti-disturbance control problem of S-MJSs with multiple disturbances is studied.Two types of disturbances are studied,one is harmonic disturbance with unknown frequency,and the other is energy bounded disturbance.In order to estimate the multiple disturbances in the system,a new composite control scheme combining feed-forward compensation of high-order disturbance observer and controller is proposed to suppress and attenuate the above two types of disturbances,respectively.In addition,sufficient conditions are given to ensure the stochastic stability of the system and satisfy the performance index.Finally,a new model of a two-degree-of-freedom(2-DOF)manipulator affected by two disturbances is constructed,and the effectiveness of the proposed scheme is verified based on this model.(3)The adaptive event-triggered composite anti-disturbance control problem of SMJSs with multiple disturbances and unknown nonlinear is studied.In order to reduce the update times of the controller and reduce the waste of channel resources,an adaptive event-triggered mechanism is introduced to avoid Zeno behavior more effectively.In order to estimate the unknown harmonic disturbance,a control based on a novel highorder disturbance observer is designed to compensate the disturbance.Based on the eventtriggered algorithm,an adaptive event-triggered composite controller is designed to reduce the computational load while stabilizing the closed-loop system.Finally,a robot example is used to verify the feasibility of the scheme.