Stochastic Integration Based Estimator: Robust Design and Stone Soup Implementation

TL;DR

This paper introduces a robust stochastic integration filter for nonlinear state estimation, implementing it within the Stone Soup framework and comparing its performance to traditional Kalman filters.

Contribution

It presents a novel stochastic integration filter and its efficient implementation in Python and MATLAB, enhancing nonlinear state estimation accuracy and robustness.

Findings

The stochastic integration filter outperforms extended and unscented Kalman filters in tracking scenarios.

Efficient square-root algorithms are developed for multi-step prediction and smoothing.

The implementation demonstrates practical applicability within the Stone Soup framework.

Abstract

This paper deals with state estimation of nonlinear stochastic dynamic models. In particular, the stochastic integration rule, which provides asymptotically unbiased estimates of the moments of nonlinearly transformed Gaussian random variables, is reviewed together with the recently introduced stochastic integration filter (SIF). Using SIF, the respective multi-step prediction and smoothing algorithms are developed in full and efficient square-root form. The stochastic-integration-rule-based algorithms are implemented in Python (within the Stone Soup framework) and in MATLAB and are numerically evaluated and compared with the well-known unscented and extended Kalman filters using the Stone Soup defined tracking scenario.