Benefits of acceleration measurement in velocity estimation and motion control

The purpose of this work is to investigate the performance of the kinematic Kalman filter (KKF), an accelerometer-assisted velocity estimation method. The most attractive feature of KKF is that it is immune to modeling uncertainties and parameter variations. Conventional velocity estimation methods are analyzed and compared to KKF. It is shown that KKF can perform very well even with low-resolution encoders. A reliable and cost-effective method of measuring angular acceleration is suggested using MEMS accelerometers. Experimental results with low-resolution encoders confirm the superiority of KKF.