Cooperative Neuro-Adaptive Fault-Tolerant Tracking Control for a Train Platoon Under Actuator Saturation.

This study investigates the cooperative fault-tolerant tracking control problem of a train platoon in the presence of actuator failure, saturation limits, and unknown operating resistance. To construct a cooperative tracking control scheme, the proposed distributed controller is designed according to adjacent trains' state information, including position, velocity, and acceleration. Radial basis function neural networks are introduced to tackle uncertain operational resistance, and the weight values of the neural network are updated online. Actuator failures with partial loss of effectiveness are estimated and compensated using adaptive methods. Another neuro-adaptive fault-tolerant control law is proposed to further solve the actuator saturation problem. The developed tracking controller remains fault-tolerant even under saturation limits while ensuring the stability of the closed-loop system. It is demonstrated using Lyapunov stability analysis that uniform ultimate boundedness is guaranteed for all signals of the closed-loop system. Numerical simulation results illustrate the effectiveness of the proposed control scheme.