Perception Matters: Enhancing Embodied AI with Uncertainty-Aware Semantic Segmentation

TL;DR

This paper improves embodied AI's object search capabilities by integrating calibrated uncertainty into perception models, enhancing decision-making in sequential tasks without extra training.

Contribution

It introduces uncertainty-aware perception calibration and aggregation methods that can be applied to existing models, addressing overconfidence issues in noisy perception during search tasks.

Findings

Calibrated uncertainties improve search accuracy.

Aggregation methods enhance perception robustness.

Uncertainty-aware models outperform baseline approaches.

Abstract

Embodied AI has made significant progress acting in unexplored environments. However, tasks such as object search have largely focused on efficient policy learning. In this work, we identify several gaps in current search methods: They largely focus on dated perception models, neglect temporal aggregation, and transfer from ground truth directly to noisy perception at test time, without accounting for the resulting overconfidence in the perceived state. We address the identified problems through calibrated perception probabilities and uncertainty across aggregation and found decisions, thereby adapting the models for sequential tasks. The resulting methods can be directly integrated with pretrained models across a wide family of existing search approaches at no additional training cost. We perform extensive evaluations of aggregation methods across both different semantic perception models and policies, confirming the importance of calibrated uncertainties in both the aggregation and found decisions. We make the code and trained models available at https://semantic-search.cs.uni-freiburg.de.