Observer-based sensor fault detection and identification with application to vehicle lateral control

Vehicle lateral control, a vital subsystem of Automated Highway Systems (AHS), acquires lateral position through two sets of on-board magnetometers. Accommodating magnetometer failures is a crucial task for the safety of AHS. Static relations between the two sets of sensors, however, are insufficient for determining the faulty set. In this paper we propose an observer-based method to deal with this problem. The proposed approach can detect and identify magnetometer failures right after failures took place. Then the output of the faulty sensor is reconstructed from the output of the healthy sensor. Therefore the same controller can stabilize the vehicle lateral control system in both normal case and sensor failure case. Simulations demonstrate that the vehicle maintains acceptable performance after either set of magnetometers has failed.