Active Fault Tolerant Torque Distribution Control of 4 in-wheel motors Electric Vehicles based on Kalman Filter approach

摘要

In this study, a fault detection algorithm based on Kalman filter approach based on the tire rotational dynamics is proposed to identify bearing fault in wheels of four in-wheel motor electric vehicles. If one wheel with partial fault is detected, feedforward compensation of bearing damping torque is added to the wheel motor torque distributed by the control allocation which integrated the MTTE-Maximum Transmissible Torque Estimator. If the one wheel with partial fault is detected, then feedforward compensation of the bearing damping torque is added to the wheel motor torque distributed by the control allocation. For the complete fault, the control allocation is necessary to be reconstructed. Other remaining normal wheels will be used for torque distribution. The bearing fault detection is faster than our previous study, because of applying Kalman filter approach in identification. That leads to improvement of lateral stability performances which are demonstrated for double lane change in simulation.