A Reset State Estimator for Linear Systems to Suppress Sensor Quantization Effects

Abstract This paper presents a reset state estimator to improve the position estimation for motion control systems with sensor quantization. The reset scheme is guided by the idea that the actual output is known exactly to be at the mid-point of the two consecutive quantizer levels and is within the range of a quantizer level bounded by half of quantization step size. Hence, using this information to update the estimated state can give a better estimation under the influence of disturbance and quantization noise. We also show that the reset scheme will not destroy the stability of a baseline estimator system. The reset state estimator is applied to a linear motor control system with an optical encoder. Simulation and experiment demonstrate that the reset state estimator can achieve smaller position estimation error and more accurate tracking accuracy than those of a standard state estimator.