Maneuvering control of stochastic nonlinear systems with unknown covariance noise

Abstract

The maneuvering problem for nonlinear systems under stochastic disturbances is investigated in this paper. Firstly, the maneuvering control objectives in their stochastic version are described in the sense of moment with tunable design parameters. Then, quartic Lyapunov functions of stabilizing errors are adopted to deal with the unknown covariance noise. Based on the adaptive law and the filter-gradient update law, an adaptive maneuvering controller is designed by the backstepping technique, which makes the closed-loop system is exponentially practically stable in mean square. Furthermore, both the path tracking error and the velocity assignment error converge to neighborhoods of zero, and the radius of these neighborhoods can be adjusted arbitrarily small by tuning independent parameters. Finally, to demonstrate the controller's effectiveness in handling unknown covariance and ensuring the practical stability of the closed-loop system, simulations of the mobile robot system in stochastic environments are conducted with various design parameters and covariance settings.