Adaptive Estimation for Brushless DC Motor Actuators in Electric Power Assisted Steering

A non-dynamic motor inverse model is used to achieve torque control of the motor in an automotive electric power steering system. Variations of the motor parameters directly impart inaccuracies in the motor control. Parameter estimation schemes based on the non-dynamic motor inverse model have been investigated in the past. This paper proposes an improved adaptive control scheme by approximating motor electrical dynamics and correcting for the back EMF compensation error caused by the velocity sampling delay. Comparison of simulations confirms a lower bound of error of the estimated parameter and faster adaptation with the proposed scheme compared to previously employed methods.