Consensus Control For Several Classes Of Nonlinear Stochastic Multi-agent Systems In Networked Environment

As we all know,multi-agent systems(MAS)are a collection of multiple agents,and its goal is to build large and complex systems into small,easy-to-manage systems.At present,the application research and theoretical models of multi-agent systems can be divided into two categories: coverage coordination(exploration,search and rescue,etc.)and motion coordination(team formation,consistency,etc.).Consensus refers to the fact that all agents in multi-agent systems are in close proximity to each other over time.The consensus problem is the basis of cooperative coordination control between agents in the control problem of multi-agents.How to effectively control the system based on the available measurement information has important theoretical and practical significance.In this paper,some new output feedback control methods are proposed for several kinds of nonlinear stochastic multi-agent systems by using linear matrix inequalities.The specific work is as follows:Firstly,in this paper,we handle the non-fragile consensus control problem for a class of discrete time-varying nonlinear multi-agent systems with uniform quantization and randomly occurring deception attacks.The quantization error is described by a noise distributed uniformly in certain interval and the phenomenon of the randomly occurring deception attacks is depicted by the Bernoulli distributed random variables with known probabilities.Here,the measurement output utilized in the controller side is not only from a single agent itself,but also from its adjacent agents,where the corresponding transmissions are carried out under a given topology.In particular,the controller gain perturbations are characterized by the multiplicative noises.We focus on the design of non-fragile time-varying output feedback controller such that,when the randomly occurring deception attacks,uniform quantization effects as well as controller gain perturbations exist simultaneously,the desired consensus performance is guaranteed by designing the controller parameters.Accordingly,a sufficient condition is derived to ensure the existence of the desired control scheme,where the feasibility of conducted problem can be tested by solving a set of matrix inequalities.Moreover,the optimal consensus performance is addressed by employing an optimization problem.Finally,we utilize the simulations to illustrate the feasibility of the proposed output feedback control strategy.Secondly,the event-triggered mean square consensus control problem for a class of discrete time-varying nonlinear multi-agent systems with missing measurement and randomly quantization is discussed.The randomly occurring quantization phenomenon is described by a Bernoulli distribution random variable with a known probability.The quantization error is described by the noise evenly distributed over a certain interval,and an event-triggering mechanism is adopted.The control input of each agent is updated only when the trigger condition is violated.We focus on the design of time-varying output feedback controller and trigger condition threshold such that,when the randomly occurring quantization phenomena,missing measurements as well as event-triggered exist simultaneously,the desired consensus performance is guaranteed by designing the controller parameters.A sufficient condition is derived to ensure the existence of the required control scheme by using linear matrix inequalities.In addition,the system achieves the best consensus performance through two optimization problems.Finally,we use simulation to illustrate the feasibility of the proposed output feedback control strategy.Thirdly,the non-fragile robust consensus control problem is studied for a class of nonlinear multi-agent systems with random one-step time delay and sensor saturation.The controller gain perturbations are characterized by the multiplicative noises.Then,a non-fragile output feedback controller is designed using the available measurement information.Nevertheless,based on the stochastic analysis method and mathematical induction approach,the controller parameters are designed to ensure that the multi-agent systems achieve the desired consensus performance.At the same time,a sufficient condition is derived to ensure the existence of the required control scheme by using linear matrix inequalities.Finally,we utilize the simulations to illustrate the feasibility of the proposed output feedback control strategy.