Least mean square method for LVDT signals processing

Abstract

This paper describes an innovative approach for position, velocity and acceleration estimation using traditional inductive sensors such as LVDTs. These transducers have been considered thanks to their diffusions and performances; however, the same algorithm can be applied to any other kind of AC-excited sensor. It overcomes typical limits of traditional processing techniques, such as coherent demodulation or spectral analysis, implementing a least mean squares estimation of interesting variables. Some preliminary simulations have been conducted to evaluate algorithm functionalities. A working prototype, designed around a low-cost DSP from TI, has been realized to perform real world tests.