RBFNN-Based Data-Driven Fast Terminal Sliding Mode Control of Nonlinear Multiagent Systems With Application to Subway Trains

This paper aims to develop the RBFNN-based data-driven fast terminal sliding mode control for nonlinear multiagent systems (MASs) subjected to unknown uncertainties and external disturbances. First, a multiagent system with unknown uncertainties and external disturbances is transformed into a full-form dynamic linearization (FFDL) data model. Then, the RBFNN algorithm is utilized to compensate for the complex unknown nonlinear terms encountered by unknown uncertainties and external disturbances, and the data-driven fast terminal sliding mode control is designed to realize the finite-time consensus tracking and enhance the robustness of MASs. Finally, a simulation test applied to subway trains is conducted to demonstrate that the proposed RBFNN-based data-driven FTSMC algorithm is characterized by high tracking accuracy.