| Cooperative control of heterogeneous multi-agent systems has become an important research direction in the field of unmanned cluster coordination technology.Multiple het-erogeneous agents work together with each other to complete the complex tasks.This cooperative control method has been applied in a variety of areas,such as unmanned co-ordination,smart grids,and cyber-physical systems.In an increasingly complex com-munications environment,a single independent communication network is struggling to meet the demands of collaborative control.As a result,clustered communication networks emerge and gradually evolve into a network topology characterized by high robustness and scalability.However,for the clustered communication networks,cooperative control of heterogeneous multi-agent systems faces a number of the novel challenges and obstacles.With regard to the output tracking problem of clustered heterogeneous multi-agent sys-tems,the work mainly addresses the following four aspects:1.The output consensus problem is investigated for the clustered heterogeneous multi-agent systems.Using the graph theory,a clustered network model with one single leader is established.Based on this,a hybrid communication mechanism is developed that combines the intermittent communication between clusters and the continuous com-munication within each cluster.According to the above communication mechanism,a distributed hybrid control strategy under the state observer is proposed to achieve output tracking of heterogeneous multi-agent systems.In addition,using the Lyapunov function method,intermittent control methods,and output regulation theory,the exponential stabil-ity of the closed-loop system is analyzed and the dwell-time conditions of output tracking are obtained.2.The output containment problem is studied for heterogeneous multi-agent systems with uncertain information over the clustered networks.First,a novel clustered network model with multiple leaders is established using the graph theory,where the hybrid com-munication mechanism involves the non-periodic intermittent communication between clusters and continuous communication within each cluster.For the clustered heteroge-neous multi-agent systems,a distributed adaptive estimation algorithm is designed to es-timate the unknown dynamics of the leaders.Based on this,a distributed adaptive hybrid control strategy under the modified intermittent control method is designed for each agent to achieve output containment.By means of theoretical analysis,the ratio between the control interval and the interval without control is obtained in order to guarantee the out-put containment.3.The output containment problem is studied for clustered heterogeneous multi-agent systems with uncertain cyber-attacks.In the case of the cluster network being subjected to uncertain cyber-attacks,an adaptive observer is proposed for each hetero-geneous agent to estimate the unknown states and external attack signals.Furthermore,a distributed hybrid control strategy under the adaptive observer to ensure the clustered het-erogeneous multi-agent systems can still achieve tracking control and containment when subjected to network attacks.4.The output containment problem is investigated for clustered heterogeneous multi-agent systems with time constraints.In the case of the clustered networks with multiple leaders,two time-constrained distributed hybrid control strategies are designed separately.In the case of finite-time control,by introducing an internal model,a distributed finite-time hybrid control strategy is designed to achieve output containment.Furthermore,to avoid the dependence of convergence time on the initial state of the agents,a distributed prescribed-time hybrid control strategy is proposed based on the time-varying scaling function and the internal model.In addition,the exponential stability of the closed-loop system is guaranteed and analyzed using the Lyapunov function method,which derives the sufficient condition of prescribed-time containment. |
PDF Full Text Request