Distributed Containment Control for Nonlinear HOFA-MASs

This work provides the distributed containment control protocols for nonlinear high-order fully actuated multi-agent systems (HOFA-MASs) under a directed communication topology graph. Based on the nonlinear HOFA model of the system, the closed-loop system can be always converted to a constant linear one with a desired eigenstructure by simply designing a controller. A reference generator is constructed for each follower by using the state information between its neighbors, then constructing the coordinate transformation so that each follower's output can track the corresponding virtual reference signal which is confined to the convex hull generated by the leaders. By using the radial basis function (RBF) neural networks (NNs) technique and the Lyapunov theorem of stability, the distributed controllers are designed for nonlinear HOFA-MASs with known nonlinear functions, nonlinear functions with unknown parameterizations and completely unknown nonlinear functions, which can guarantee that the outputs of followers can enter into the convex hull. Finally, the simulation results on robotic system demonstrate the availability of the designed protocols.