Look No Further: Adapting the Localization Sensory Window to the Temporal Characteristics of the Environment

TL;DR

This paper introduces an adaptive method for setting the sensory window length in localization algorithms, optimizing context use based on environmental cues to improve accuracy and reduce travel without localization.

Contribution

It proposes a general environment-driven approach for dynamically adjusting the spatiotemporal window length in localization systems, eliminating the need for deployment-time tuning.

Findings

Reduces maximum travel distance without localization.

Achieves competitive localization accuracy.

Works with visual and Wi-Fi sensors in various traversal orders.

Abstract

Many localization algorithms use a spatiotemporal window of sensory information in order to recognize spatial locations, and the length of this window is often a sensitive parameter that must be tuned to the specifics of the application. This letter presents a general method for environment-driven variation of the length of the spatiotemporal window based on searching for the most significant localization hypothesis, to use as much context as is appropriate but not more. We evaluate this approach on benchmark datasets using visual and Wi-Fi sensor modalities and a variety of sensory comparison front-ends under in-order and out-of-order traversals of the environment. Our results show that the system greatly reduces the maximum distance traveled without localization compared to a fixed-length approach while achieving competitive localization accuracy, and our proposed method achieves this performance without deployment-time tuning.

Figures

Full text

See pages 1-last of paper.pdf