Robust Load-Torque Estimation for DC Motor without Torque Sensor

Abstract

External load-torque estimation for electrical motor is important in order to improve control performance as well as obtaining information about interaction with the environment. This paper presents a performance and robustness comparison among three different types of algorithms for the estimation of the external load-torque for low-power DC motors. The Kalman filter is presented as the standard estimation technique and it is compared to the H∞filter and the Super-Twisting Sliding Mode Observer (STSMO). The algorithms are based on the position and speed measurements of the rotor.