Observer design for sensor and actuator fault estimation applied to polynomial LPV systems: a riderless bicycle study case

The present paper addresses an observer design for fault estimation applied to polynomial linear parameter varying (LPV) systems. The main contribution corresponds to a design to provide the estimation of both actuator and sensor faults acting on the system, modelled as additive faults taking place at any time and simultaneously, inclusively. The observer design includes a boundary factor in order to guarantee the estimation error convergence to zero despite the fault occurrence. Simulation results showing the effectiveness of the proposed observer by considering a riderless bicycle LPV model, which depends on the translational velocity of the vehicle, are provided.