Adaptive neural control for unmanned surface vessels with asymmetric full-state constraints

This paper is devoted to solving the tracking control issue for unmanned surface vessels (USVs) subject to asymmetric full-state constraints, in which the uncertainties are explicitly considered. Uncertainties of USVs is handled by neural approximation approach. A new asymmetric barrier Lyapunov function (ABLF) is first established, and it not only solves the problem of state constraints but also avoids the conservative selection of state initial values. Furthermore, a novel adaptive neural backstepping design scheme is put forward by utilizing ABLF and hyperbolic tangent function approaches. The proposed scheme guarantees that all the state constraints are never violated and the system output follows the given reference signal well during the whole control procedure. Finally, simulation results have illustrated the efficacy of our method.