Spot the Difference: A Novel Task for Embodied Agents in Changing Environments

TL;DR

This paper introduces a new task called 'Spot the Difference' for embodied AI agents, focusing on detecting environmental changes using outdated maps, and presents a dataset and exploration policy that outperform existing models.

Contribution

The paper proposes a novel environment change detection task, creates a new dataset of occupancy maps, and develops an exploration policy that leverages prior knowledge for improved change detection.

Findings

The proposed architecture outperforms state-of-the-art exploration models.

The dataset enables training and evaluation of change detection in embodied AI.

The exploration policy effectively identifies scene changes faster.

Abstract

Embodied AI is a recent research area that aims at creating intelligent agents that can move and operate inside an environment. Existing approaches in this field demand the agents to act in completely new and unexplored scenes. However, this setting is far from realistic use cases that instead require executing multiple tasks in the same environment. Even if the environment changes over time, the agent could still count on its global knowledge about the scene while trying to adapt its internal representation to the current state of the environment. To make a step towards this setting, we propose Spot the Difference: a novel task for Embodied AI where the agent has access to an outdated map of the environment and needs to recover the correct layout in a fixed time budget. To this end, we collect a new dataset of occupancy maps starting from existing datasets of 3D spaces and generating a number of possible layouts for a single environment. This dataset can be employed in the popular Habitat simulator and is fully compliant with existing methods that employ reconstructed occupancy maps during navigation. Furthermore, we propose an exploration policy that can take advantage of previous knowledge of the environment and identify changes in the scene faster and more effectively than existing agents. Experimental results show that the proposed architecture outperforms existing state-of-the-art models for exploration on this new setting.