Event-Triggered Control For Multi-Agent Systems

In practical applications, controllers are in general implemented on digital plat-forms, and control laws can only be updated at discrete instances of time. Currently, the most common way is periodic sampling, in which controller updates are determined by a constant sampling period. However, the sampling time of periodic sampling system might be conservation in terms of the number of control updates, since the constant sam-pling period has to guarantee stability in the worst-case scenario. With the progress of technology, agent may equip with an embedded microprocessor, which will be responsi-ble for exchanging information with neighboring agents and actuating the individual agent controller updates. The individual agent has limited energy, which requires the system to reduce energy consumption as much as possible. The controllers update when some certain conditions are satisfied with event-triggered control strategy, and the energy can be saved by this way. Event-triggered multi-agent system not only guarantees the control performance of system to some extent, but also reduces the number of controller updates greatly. Event-triggered formation control for multi-agent systems is studied in this thesis. The main results and contributions of this thesis are as follows.This paper mainly investigates the event-triggered control for multi-agent systems with the problem of packet losses where both the first-order and the second-order neigh-bors’ information are used. The problem is formulated under the sampled-data frame-work for the second-order and the high-order discrete time agent dynamics. The commu-nication graph is undirected and the loss of data across each communication link occurs at certain probability, which is governed by a Bernoulli process. The mean square stabil-ity of the closed loop multi-agent systems is analyzes based on the Lyapunov functional method. It is found that numerical examples are given to demonstrate the effectiveness of the proposed methods. Then the distributed consensus speed up by using the se-cond-order neighbors’ information when packet losses.Based on the analysis of first-order integrator system, the event-triggered control strategy for heterogeneous multi-agent systems composed of two kinds of agents differed by their dynamics is also investigated in this thesis. In addition, time-varying delay is taken account as well. The distributed control scheme is presented at first. Then, the consensus problem of heterogeneous multi-agent systems is solved based on a proposed control law under the Lyapunov functional method.