A Nonlinear Observer Based Analytical Redundancy for Predictive Fault Tolerant Control of a Steer-by-Wire System

In this paper, a nonlinear sliding mode observer, along with a long range linear predictor, is presented for fault tolerant control of a steer by wire system. The long-range predictor is based on Diophantine identity aimed at improving the fault detection efficiency. The overall predictive fault tolerant control strategy was then implemented and validated on a steer by a wire hardware in loop bench. The experimental results show that the overall robustness of the steer by wire system was not sacrificed through the usage of analytical redundancy for sensors along with the designed fault detection, isolation, and identification algorithm. Moreover, the experimental results indicated that the fault detection speed is improved using the proposed analytical redundancy-based algorithms for both attenuating and amplifying type faults. The proposed fault detection algorithm was also found to be robust against a wide range of fault types.