A GoA4 Based Distributed Adaptive Control for multi-body high-speed trains with passive inter-cars couplers

This paper tackles the problem of controlling the motion of a fully-autonomous nonlinear High-Speed Train (HST) composed of several power cars, physically connected via passive couplers. Performing acceleration and deceleration manoeuvres at high speed could generate higher longitudinal inter-cars coupling forces, which may cause possible shocks to the couplers and compromise their mechanical structures, as well as safety, during the travel. To address this issue, we extend the Grade of Automation 4 (GoA4) controller functionalities proposed in our previous work for a single car to the multi-body framework by proposing a novel distributed adaptive control protocol which is able to guarantee that the whole nonlinear HST moves according to the imposed journey profile. Specifically, while the first car is driven via our previously proposed GoA4 controller, the distributed adaptive control aims to pilot the behaviour of the connected nonlinear power cars to reduce inter-cars shocking phenomena. Moreover, the proposed distributed strategy does not require any knowledge or estimation of the nonlinear car dynamics. The stability of the closed-loop HST dynamics, along with the boundedness of the adaptive signals, is proven by leveraging the Lyapunov approach and Barbalat Lemma. The derived stability conditions also allow us to find some proper tuning conditions for the control gains. Simulation results, carried out on a realistic railway track scenario, confirm the effectiveness and the benefits of the proposed control architecture in guaranteeing a smooth and safe inter-car motion.