Estimation of Higher-Order State Variables by Polynomial Approximation Considering Quantization Step Size of Encoder and the Implementation on FPGA

Estimation of higher-order state variables such as velocity and acceleration requires high accuracy and wide bandwidth in position measurement in motion control. The purpose of this study is to resolve the trade-off between the accuracy and delay in the estimation of higher-order state variables from an encoder. This paper proposes a novel estimation method using polynomial fitting that takes the quantization step size of an encoder into account. The developed approach makes higher accuracy of the estimation with the same estimation delay. The improvement of the accuracy is validated and the possibility of the real-time use of the proposed method in motion control is validated using FPGA.