Research On Event-triggered Control Based On Hybrid System Approach

Equipments in networked control systems usually communicate via digital communication networks.Due to many benefits they provide,including greater flexibility,ease of installation and low cost,these systems have drawn a lot of attention in recent years.Great developments have been obtained in many industrial fields,such as intelligent transportation,multiple drones and remote surgery.However,wireless communication inevitably brings the issue of limited communication resources,which leads to new challenges that the precious communication resources should be carefully used.Hybrid system is a kind of system which combines flow dynamics and jump dynamics.The discontinuity induced by communication in networked control systems can be perfectly modelled by the jump dynamics of the hybrid system.The idea of the hybrid system approach is to first build a hybrid system model for the networked control system,and then utilize Lyapunov-based tools to analyze the stability property.Unfortunately,there are still many uncharted territories.This paper aims at exploring some challenging problems based on the hybrid system approach.Specifically,this paper consists of two parts: dynamic-quantizer-based periodic event-triggered control of nonlinear systems,and a co-design approach to distributed dynamic event-triggered communication and control for multi-agent consensus.The first part investigates the event-triggered quantized control for nonlinear systems with external disturbances by exploiting both temporal and spatial aspects.In order to cope with the issues that inhibit the practical implementation such as the continuous monitoring of the event-triggering condition and finite-bit data at each transmission,the periodic event-triggering mechanism and the dynamic quantizer are employed respectively.Two configurations of the event-based quantized controller,i.e.,quantization after sampling and quantization before sampling,are discussed for the wider utilization of the quantized control.By exploiting the hybrid system approach,the resulting control systems in the two configurations can be modeled and guaranteed to possess input-to-state stability property of a size-adjustable set around the origin.The effectiveness of the proposed approach is illustrated in numerical examples.The second part investigates the output-based event-triggered communication and control for linear multi-agent consensus under directed graph with external disturbances.In order to further reduce the use of the system resources,we consider the scenario in which both the communication among agents and the controller update are determined by the event-triggering mechanisms.To simultaneously guarantee the significant properties including the asymptotic consensus,and strong Zeno-freeness(strictly positive interevent times),a novel distributed dynamic event-triggered protocol is proposed.Unlike most-existing works in which the control gain is previously decided,a co-design method is proposed to design the control gain,the observer gain and the event-triggering mechanisms altogether in terms of solving a linear matrix inequality optimization problem.Based on the hybrid system framework,a novel decoupled model is constructed for the distributed closed-loop system,and the consensus is achieved asymptotically.Finally,a numerical example is illustrated to verify the effectiveness of the proposed methods.