Robust Filtering and Smoothing with Gaussian Processes
This work addresses robustness issues in filtering and smoothing for applications like robotics and control, but it appears incremental as it builds on existing GP dynamic systems.
The authors tackled robust Bayesian filtering and smoothing in nonlinear stochastic dynamic systems using Gaussian process models for both transition and measurement functions, demonstrating robustness where other state-of-the-art Gaussian filters and smoothers fail in numerical evaluations.
We propose a principled algorithm for robust Bayesian filtering and smoothing in nonlinear stochastic dynamic systems when both the transition function and the measurement function are described by non-parametric Gaussian process (GP) models. GPs are gaining increasing importance in signal processing, machine learning, robotics, and control for representing unknown system functions by posterior probability distributions. This modern way of "system identification" is more robust than finding point estimates of a parametric function representation. In this article, we present a principled algorithm for robust analytic smoothing in GP dynamic systems, which are increasingly used in robotics and control. Our numerical evaluations demonstrate the robustness of the proposed approach in situations where other state-of-the-art Gaussian filters and smoothers can fail.