Adaptive windowing discrete-time velocity estimation techniques: Application to haptic interfaces

A method is described to estimate velocity from discrete and quantized position samples via adaptive windowing. It addresses the shortcomings of previously known methods which necessitate tradeoffs between noise reduction, control delay, estimate accuracy, reliability, computational load, transient preservation, and which cause difficulties with tuning. The method is optimal in the sense that it minimizes the velocity error variance while maximizes the accuracy of the estimates. The design of the estimator requires the selection of only one parameter, namely a bound on the noise. Simulation and experimental results are presented.