Cooperative Control Of Multi-Agent Systems With Multi-Source Disturbances

Many biological cluster phenomena in nature have aroused the interest of many researchers.People try to reveal the inherent rules of these phenomena from the theoretical point of view.Through understanding their essence,we can better serve human society.The multi-agent systems have been emerging,which is to complete a complex work by interaction of each agent.Nowadays,the multi-agent systems have applied widely in many fields,such as UAV cooperative fight,Underwater Vehicle in the sea bed exploitation,satellite positioning and attitude adjustment,logistics warehousing intelligent scheduling intelligent vehicle driving and other engineering fields.In practical engineering,due to unmolded dynamics,parameter uncertainty,wind,noise,temperature and other external factors,the steady-state performance of multi-agent control system is seriously affected.In this paper,an active anti-disturbance control method is used to study the distributed cooperative control of dynamic multi-agent systems in complex environments.The main contents and achievements include:(1)Consensus for leader-follower multi-agent systems with multi-source disturbances.By converting mismatched disturbance into matched disturbance through model transformation,an active disturbance observer is designed to estimate the disturbance of each agent.Then,a composite distributed control protocol with the disturbance feedforward compensation is proposed.By using the theory of modern control and linear matrix inequality,the consensus problem of multi-agent systems with disturbances under undirected topology and directed topology are discussed.(2)Robust H? consensus of multi-agent systems with mismatched disturbances and parameter uncertainties.Consider the mismatched disturbance in velocity channel and matched disturbance in input channel,a disturbance observer is developed such that the disturbances in multi-agent systems can be estimated and compensated directly.Then a novel composite disturbance observer-based control law is presented to study the robust consensus of multi-agent systems with mismatched disturbances.Supposing the case of the parameter uncertainty in the system,robust consensus of multi-agent systems with mismatched disturbances and parameter uncertainties is discussed based on matrix inequality theory.(3)Finite-time global consensus for multi-agent systems with mismatched disturbances.First of all,a nonlinear finite-time disturbance observer is constructed for each agent to estimate its states and mismatched disturbances.Second,dynamic sliding-mode surfaces are designed by employing its own and neighbors' disturbance estimates.Then,based on the designed surfaces,the discontinuous or continuous integral sliding mode control protocols are respectively proposed to guarantee the active anti-disturbance consensus for multi-agent systems with disturbances in finite time.(4)Robust flocking motion of second-order multi-agent systems with mismatched disturbances.For the mismatched disturbances,a disturbance observer is designed to estimate the disturbances.Then based on artificial potential function,a composite distributed control protocol with feed-forward compensation term is proposed.By using Lyapunov stability and input-to-state stability(ISS)theory,the flocking motion of multi-agent systems with mismatched disturbances is studied.Consider the boundedness of the disturbance,the sufficient condition of the robust consensus of the multi-agent systems is obtained based on robust H? control.