Data-Driven Measurement Models for Active Localization in Sparse Environments

TL;DR

This paper presents a data-driven approach for environment exploration and object localization using ergodic exploration, enabling robots to efficiently build measurement models and localize objects in sparse environments.

Contribution

The authors introduce a two-stage algorithm combining ergodic exploration and information-based localization for sparse environment mapping and object detection.

Findings

Effective in identifying and localizing objects with sparse binary contact data

Visits to low-probability regions enhance information gain

Validated on tactile sensing simulations and robot experiments

Abstract

We develop an algorithm to explore an environment to generate a measurement model for use in future localization tasks. Ergodic exploration with respect to the likelihood of a particular class of measurement (e.g., a contact detection measurement in tactile sensing) enables construction of the measurement model. Exploration with respect to the information density based on the data-driven measurement model enables localization. We test the two-stage approach in simulations of tactile sensing, illustrating that the algorithm is capable of identifying and localizing objects based on sparsely distributed binary contacts. Comparisons with our method show that visiting low probability regions lead to acquisition of new information rather than increasing the likelihood of known information. Experiments with the Sphero SPRK robot validate the efficacy of this method for collision-based estimation and localization of the environment.