Development of Precise Encoder Edge-Based State Estimation for Motors

Abstract

We developed and implemented an algorithm named Precise Encoder Edge-based State Estimation for Motors (PEESEM). Instead of conventional periodic encoder sensing to obtain the quantized motor position, we detect the encoder edge's time and value precisely. Then, we use the edge-time Kalman filter (ETKF) containing predictions at edge time and periodic sampling time, and update at edge time. Only the first edge in each sampling interval is utilized to reduce the computation time at high motor speed. The proposed algorithm guarantees far more accurate state estimation with low encoder resolution and uncertainty on motor parameters. Performance of the proposed algorithm is validated through simulations and implementation on a two-wheeled mobile robot (TMR)