Event-Trigger-Driven Active Disturbance Rejection Control For Nonlinear System In Pure Feedback Form

Many engineering applications in modern civil use and national defense have pure feedback nonlinear system forms,such as manipulator and unmanned aerial vehicle.An actual cooperation site subjects to disturbance inevitably,such as model uncertainty,external disturbance.In the circumstance of limited communication resources,it is of great theoretical and engineering significance to study disturbance rejection control with an event-triggered manner for pure feedback nonlinear system.In this paper,the control issue for a class of uncertain pure feedback nonlinear systems is investigated.Under the framework of backstepping method,a lumped uncertainty is regarded as new system state variable,and active disturbance rejection control and event-triggered mechanism are combined to overcome unnecessary network transmission under time-driven control.The active disturbance rejection control problem of a class of uncertain pure feedback nonlinear systems is studied.The main work is summarized as three aspects.First,a dual-terminal triggered control scheme with an event-triggered extended state observer and a trigger-driven input is proposed for a tracking control problem of uncertain affine nonlinear system with unknown external disturbance and unknown control direction.Combined with backstepping technique and active disturbance rejection control,by a Nussbaum-type function and an adaptive estimation parameter,an output feedback tracking control law based on event-triggered extended state observer and trigger-driven input is designed.The event-triggered extended state observer reconstructs the system states and compensates the effects of uncertain and disturbance in real time.Compared with the continuous time extended state observer,the event-triggered one removes the assumption that the output information is continuously available and reduces the number of data transmission between the system and the observer.The core idea of trigger-driven input is that the control signal will be updated immediately once the preset triggered condition is violated.The trigger-driven input reduces the communication times between the controller and the actuator regulated by an event-triggered mechanism.The stability of the closed-loop system and the boundness of all the signals in the closed-loop are analyzed by Lyapunov function,and Zeno phenomena in the system is ruled out.Second,for a tracking control issue of a class of non-affine nonlinear systems with uncertain,an event-triggered tracking control scheme with predefined performance is proposed.The non-affine system is transformed into an affine one by combination use of implicit function theorem,mean value theorem and active disturbance rejection control principle.Event-triggered manner reduces the communications times from controller to actuator.An extended state observer is designed to estimate and compensate system uncertain in real time.A tracking differentiator eliminates the complex derivation computation brought by backstepping method.Also,system tracking performance is improved by introducing a predefined performance function.Further,Taking a loss effectiveness and stuck failure of actuators into account,we propose an adaptive event-triggered fault-tolerant control scheme with predefined performance.Unknown actuator faults are handled by compensation mechanism.A pedefined performance function is employed to improve tracking accuracy.An event-triggered mechanism with time-varying threshold reduces the communication burden from the controller to the actuator.The extended state observer and the tracking differentiator are used to reconstruct the unknown dynamics of the system and to simplify the high order derivative operation of the virtual control law,respectively.The stability of the closed-loop system and the boundedness of all the signals in the closed-loop system are proved by constructing an input-to-state stability-Lyapunov function.And the Zeno-free phenomena is proved.Third,for a issue of formation collision-avoiding control of a class of uncertain non-affine pure nonlinear multi-agent systems,with an event-triggered communication fashion,a distributed formation collision-avoiding control scheme via an improved time-varying repulsive force gain coefficient is proposed.Combined with the implicit function theorem,mean value theorem and the principle of active disturbance rejection control,the non-affine multi-agent system is transformed into affine one.The artificial potential field is introduced to solve the problem of collision avoidance between followers and obstacle avoidance.The improved time-varying repulsive force gain coefficient in the control law improves the incoordination between repulsive force and formation control variable existing in traditional artificial potential field algorithm.An extended state observer approximates the uncertain of each agent and compensates in real time.The tracking differentiator avoids the explosion phenomenon of virtual control law derivation computation.A distributed event-trigger-driven estimator is employed to reconstruct leader information,which saves the communication resources and reduces the communication burden between the followers of the whole system.With a constracted input-to-state Lyapunov candidate function and an energy function whose time derivative is positive,it is proved that the multi-agent system can realize formation control and has function of avoiding obstacles and collision,and it is also proved that the system exists Zeno-free phenomenon.