Modified Sage-Husa adaptive Kalman filter for multi-source noise mitigation in dual-comb ranging

Dual optical comb ranging has the advantages of high measurement accuracy and fast update rate; however, it is susceptible to environmental disturbances in practical applications, and its measurement information is often mixed with time-varying and non-stationary environmental noise, which results in the decrease of ranging accuracy. In this paper, we proposed a modified Sage-Husa adaptive Kalman filtering (SHAKF) algorithm combining residuals and sliding windows for multi-source noise reduction in dual-comb absolute distance measurement systems. In addition, the VMD-SWT algorithm is introduced to digitally process the interferometric signals. The algorithm retains the Sage-Husa filter forgetting factor, introduces exponentially weighted updating of Q and R to enhance the real-time performance of parameter estimation, presenting a sliding window strategy to deal with sudden change noise while adaptively adjusting the parameters according to the noise. Our method has been verified to effectively suppress multiple types of noise, improving the ranging accuracy by an order of magnitude to the micrometer level and reducing both the mean absolute error (MAE) and root mean square error (RMSE) by more than 63% across five measurement sets.