Adaptive Output Feedback Containment Control for Stochastic Multi-agent Systems with Input Saturation

In this paper, a finite-time adaptive containment control method is proposed for non-strict feedback stochastic multi-agent systems (SMASs) with input saturation. For the first time, we propose the definition of the containment control in the sense of probability, and use Chebyshev inequality in stability analysis to prove that controller can satisfy the definition with a large probability. The unknown state is estimated by high-gain observer, and the unknown continuous function is approximated by using radial basis function neural networks (RBFNNs). Based on stochastic finite-time theory, we combine dynamic surface control (DSC) and nonlinear command-filter to simplify the controller design, and cancel the assumption that the filtering error is bounded. Finally, the stability analysis and simulation experiments prove that all signals in the controlled plant are semi-globally finite-time stability in probability (SGFSP).