Efficient Greedy Algorithms for Feature Selection in Robot Visual Localization

TL;DR

This work tackles real-time feature selection for robot visual localization under onboard resource limits by formulating it as a monotone submodular maximization problem. It introduces two greedy-based algorithms: a Stochastic-Greedy method that achieves near-optimal guarantees with reduced evaluations, and an Approximate-Greedy method that uses a trace surrogate to dramatically cut memory usage. Theoretical results establish an $(1 - 1/e - \varepsilon)$-type approximation for the stochastic approach and a per-feature trace characterization that justifies the surrogate-based method, including that $\operatorname{tr}(\mathbf{H}_{\star}^{f}) = 2 n_f - 3$. Collectively, these methods enable efficient, onboard feature selection for visual localization with potential applications in multi-robot settings and adaptive horizon strategies, while maintaining localization accuracy.

Abstract

Robot localization is a fundamental component of autonomous navigation in unknown environments. Among various sensing modalities, visual input from cameras plays a central role, enabling robots to estimate their position by tracking point features across image frames. However, image frames often contain a large number of features, many of which are redundant or uninformative for localization. Processing all features can introduce significant computational latency and inefficiency. This motivates the need for intelligent feature selection, identifying a subset of features that are most informative for localization over a prediction horizon. In this work, we propose two fast and memory-efficient feature selection algorithms that enable robots to actively evaluate the utility of visual features in real time. Unlike existing approaches with high computational and memory demands, the proposed methods are explicitly designed to reduce both time and memory complexity while achieving a favorable trade-off between computational efficiency and localization accuracy.