Iterative Learning Containment Control For Discrete Time Systems And Under Fading Communication

Researches on multi-agent systems(MASs)are inspired by emergency phenomena of groups in natural world.Containment control is such that multi-agent systems include multiple leaders and followers are driven into the convex hull spanned by all those leaders.Containment control is raised great concern for its widespread applications in transportation of vehicle,phase control of radar,collaborative production of processing industry,etc.In practical issues of containment control,there is a class of control problems that contain repetitive tasks and require control precision.Such control problems can be solved by iterative learning containment control(ILCC)algorithm.Iterative learning containment control has typical application scenarios.However,there are a few research results in related area.Hence researches on this issue are of momentous meaning in theoretical analyses and great value in practical applications.Firstly,iterative learning containment control laws are designed for followers described by both linear discrete systems and affine nonlinear discrete systems.A specific combination of outputs are chosen as the desired outputs of followers and control laws are designed so that all outputs errors of followers tend to zero as the iteration number increases.Eventually,precise tracking of the desired outputs in finite time of all followers is realized.The proposed control laws are proven effective by convergence analyses and numerical simulations.Secondly,iterative learning containment control laws are designed under the circumstances where transmitted signals between agents are influenced by fading communication environment.Signals rectification and decreasing step sizes along the iteration axis are introduced to suppress the influences caused by noises in fading channels.The proposed control laws can guarantee that all followers outputs in finite time interval tend to the desired outputs in the mean square sense as the iteration number increases.The desired outputs are a specific combination of outputs.Research results show that,according to the designed iterative learning containment control laws in this thesis,containment control effects of discrete time systems under both lossless and fading communication are realized.Such results are of great significance for researches in iterative learning coordination control of multi-agent systems and applications of competitive containment control.