Model free Adaptive Predictive Control with Prescribed Performance for Discrete-Time Nonlinear Multiagent Systems

This article is devoted to investigating model-free adaptive predictive control scheme based on prescribed performance for nonlinear multiagent systems (MASs). A novel tangent-type error transformation function is proposed for mapping performance indicator from constrained to unconstrained. Compared with the common logarithmic-type error transformation function, the computational complexity is reduced. Meanwhile, the convergence of distributed measurement error among agents to a predefined area is achieved. Furthermore, the proposed scheme is data-driven method based on compact form dynamic linearization without any model information. Theoretical analysis ensures the consensus of multiple agents and the analysis result is verified by simulation.