Actuator-Redundancy-Based Fault Diagnosis for Four-Wheel Independently Actuated Electric Vehicles

This paper presents a real-time actuator-redundancy-based fault diagnosis approach for four-wheel independently actuated (FWIA) in-wheel motor electric ground vehicles. The tire-road friction coefficient (TRFC), which is usually unknown, needs to be accurately estimated to calculate the in-wheel motor torque and evaluate the fault in real time. An observer is applied to each of the in-wheel motors to generate a TRFC estimate from the respective wheel. The individual estimates of the TRFC are merged using a voting scheme, which can reject the erroneous estimate from the faulty motor. Then, the resulting accurate TRFC estimate is adopted to calculate the residual for each of the in-wheel motors and to detect the possible actuator fault. Experimental results from a prototype FWIA electric ground vehicle are given to show the effectiveness of the proposed fault diagnosis method.