| The regulation of feedforward nonlinear systems is an an important research topic in the control theory of nonlinear systems.It has been drawing much attention from academic communities.On the one hand,many practical systems can be transformed into feedforward nonlinear systems by proper coordinate transformation,such as ball and beam systems with friction,cart-pendulum systems,LLC resonant circuit systems and so on.On the other hand,due to the complex physical environment,low accuracy of measuring tools,and limited modeling approaches,the mathematical models of actual systems often contain the parameter and structural uncertainties.These uncertainties inevitably affect the control performance and hinders the realization of stabilization.In this situation,it is difficult to achieve the stabilization.Therefore,the global regulation control is gradually favored by scholars,which means that,for any initial condition,the system states are bounded and converge to zero under the controller.In view of this,with the aid of timevarying gain method,this dissertation studied event-triggered control problems for uncertain feedforward nonlinear systems and output feedback consensus problems for feedforward nonlinear multi-agent systems subject to state delays,input delays and communication delays,to achieve global regulation of closed-loop systems.The main contents are summarized as follows:1.The event-triggered output feedback control problem was studied for uncertain feedforward nonlinear systems with unknown measurement sensitivity.First,an auxiliary system was constructed by the time-varying gain,which is independent of the output signal.Then,a constant gain was also introduced to handle unknown measurement sensitivity.Second,a novel triggering mechanism based on a logarithmic function was proposed to design an event-triggered output feedback controller,which achieved that system states are globally bounded and ultimately converge to zero.Furthermore,it is verified that the triggering intervals have a positive lower bound in the infinite time domain.2.The event-triggered fault-tolerant control problem was studied for highorder feedforward nonlinear systems with actuator faults.First,an adaptive law was introduced based on the time-varying gain,and a concise fault compensation scheme was proposed.Second,a triggering mechanism composed of three triggering functions was designed to determine when the system states transform to the controller,and to design fault-tolerant controller.Third,with the mathematical induction,it was proved that the event-triggered fault-tolerant controller enabled the closed-loop system to achieve global regulation,and effectively avoided Zeno behavior.3.The event-triggered output feedback consensus problem was studied for uncertain feedforward nonlinear multi-agent systems with time delays.First,a distributed observer was constructed based on sampled output information.Then,by constructing a relative triggering mechanism,an event-triggered control protocol was designed to achieve leader-follower consensus for multi-agent systems with state delays.Second,a compensator was constructed,which is independent of input delays directly.Then,two periodic event-triggered output feedback control protocols are designed to achieve leaderless consensus for multi-agent systems with uniform time-varying delays,and leader-follower consensus for multi-agent systems with non-uniform time-varying delays,respectively.The triggering intervals had a positive lower bound in the infinite time domain under the proposed triggering mechanisms.4.The output feedback scaled consensus problem was studied for feedforward nonlinear multi-agent systems with communication delays and bounded noises.When only sampling output information can be obtained,a new type of relative output variable based on the time-varying gain was introduced to construct the distributed observer and output feedback control protocol.It effectively overcame the dificulty that an agent cannot independently obtain the current output information of its neighbors or noises of communication channels.Furthermore,it achieved the scaled consensus for multi-agent systems.Notably,it is unnecessary to restrict the upper bound of the sampling period,communication delays and communication noises.Moreover,the consensus is a particular case of the scaled consensus.In summary,this dissertation designed a series of control schemes based on time-varying gain to achieve global regulation for several typical uncertain feedforward nonlinear systems.The main contributions are summarized as follows:(1)An adaptive compensation mechanism based on the time-varying gain was introduced to the event-triggered fault-tolerant controller.It effectively handled uncertain nonlinear terms and unknown parameters of actuator faults,and achieved the regulation for high-order feedforward nonlinear systems.(2)By designing triggering mechanisms based on logarithmic function and output feedback control protocols,consensus errors were globally bounded and ultimately converged to zero without continuous communication between agents.It also ensured that the triggering intervals had a uniform positive lower bound.(3)A distributed observer and a control protocol were designed based on sampled outputs by introducing a new relative output variable dependent on time-varying gain.It effectively dealt with communication delays and unknown communication noises,and then achieved the scaled consensus for multi-agent systems. |
