Cluster Control Problems In Virtual Structures And Their Numerical Method

The control of time-varying systems is an important part of modern mathematical control theory and one of the most active research fields at present.The design and research of feedback strategy is an important breakthrough direction to realize the process control of complex systems,which is widely used in military,medical and other scientific and technological fields.In particular,it can also be used to study the swarm motion in biological systems such as geese or fish.It is the key to realize cluster bionic optimal control.This thesis studies the cluster formation control system that moves jointly to a given target.The main contents include the target control of a single cluster,the coordinated movement of multiple clusters and the dynamic reorganization of clusters.The dynamic reorganization of clusters includes the increase and decrease of cluster members,the merger of clusters and the split of clusters.In addition,with the rapid development of offshore operation technology and the complex and changeable low-altitude sea environment,the thesis also gives the relevant research results of cluster control under the influence of sea wind field and tide.For the problem of cluster formation control,a complete solution based on Hamiltonian method is proposed in this thesis.In order to ensure the smooth wireless information exchange among the members of the cluster,all the members of the cluster are assumed to be in a virtual container during the whole movement.First,the virtual container orbit is constructed as the external state constraint of the cluster.On the premise of investigating the problems of collision avoidance and energy constraint among the members during the movement of the cluster,a matrix-valued function is constructed.On the basis of Hamiltonian form,the model and optimization algorithm of each cluster target control problem are given,and then the optimal control of each member in the cluster satisfying the constraint conditions is obtained.Numerical experiments and simulation verify the practical availability of the method.