Informative Planning for Worst-Case Error Minimisation in Sparse Gaussian Process Regression

TL;DR

This paper introduces a planning framework that minimizes the worst-case error in sparse Gaussian process regression by leveraging a novel entropy minimization approach, validated through simulations in 1D and 2D scenarios.

Contribution

It derives a universal worst-case error bound for sparse GP regression and proposes a posterior entropy minimization method for error reduction, outperforming traditional approaches.

Findings

Effective in minimizing deterministic error in GP regression.

Outperforms measurement entropy maximization in fixed inducing point scenarios.

Validated through simulations in 1D and 2D environments.

Abstract

We present a planning framework for minimising the deterministic worst-case error in sparse Gaussian process (GP) regression. We first derive a universal worst-case error bound for sparse GP regression with bounded noise using interpolation theory on reproducing kernel Hilbert spaces (RKHSs). By exploiting the conditional independence (CI) assumption central to sparse GP regression, we show that the worst-case error minimisation can be achieved by solving a posterior entropy minimisation problem. In turn, the posterior entropy minimisation problem is solved using a Gaussian belief space planning algorithm. We corroborate the proposed worst-case error bound in a simple 1D example, and test the planning framework in simulation for a 2D vehicle in a complex flow field. Our results demonstrate that the proposed posterior entropy minimisation approach is effective in minimising deterministic error, and outperforms the conventional measurement entropy maximisation formulation when the inducing points are fixed.