LAUSR

Acquiring the relative displacement of adjacent vehicles when designing a distributed controller for a heavy-haul train is a critical issue. As heavy-haul trains run in complex, varying environments, the relative displacement of adjacent vehicles may not be measured accurately. To address this issue, this study proposes a distributed cooperative observer to estimate the relative displacement among train vehicles. First, a mass-point elastic-coupled dynamic model is constructed to capture the relative displacement of a heavy-haul train, and the complex model is further decomposed into a series of reduced double integrators. Then, a distributed cooperative dynamic observer is designed for each reduced double integrator. State information is shared via communication among neighbouring observers to accurately estimate the relative displacements of the vehicles under system noise. The convergence of the cooperative observer error is analysed using a Riccati equation. Finally, the performance of the closed-loop heavy-haul train control system combined with the cooperative observer is rigorously evaluated using real parameter settings of Shuohuang Railway in China.