Distributed Cooperative Control of High-Speed Trains With Input Saturation and Unmodeled Dynamics

In this work, a distributed cooperative control strategy is proposed for multiple high-speed trains (MHSTs) subject to input saturation and unmodeled dynamics. To facilitate the distributed controller design, the dynamics of trains are firstly modeled as a muti-agent system (MAS) with a state-dependent directed graph. Then, the distributed control laws that are equipped with a command filter and a robust adaptive neural network are developed to achieve the consensus task among the MHSTs with predefined displacement and speed trajectories. It should be highlighted that, to reduce the complexity of the control algorithms and computational burden, we propose adaptive estimation laws to estimate the upper bound of the norm of the neural network weight vectors, instead of the weights themselves. Further, an auxiliary dynamical system (ADS) is introduced to compensate the influence of the input saturation. The convergence of the proposed controllers are analyzed rigorously by applying Lyapunov theorem, and the effectiveness of the proposed control approach is demonstrated by numerical simulations.