On Evaluation of Embodied Navigation Agents

TL;DR

This paper reviews the current state of embodied navigation agents, highlighting the need for standardized evaluation protocols and proposing benchmark scenarios to unify research efforts in this rapidly evolving field.

Contribution

It offers a consensus on empirical evaluation methods, discusses various problem statements, and introduces standard benchmarks for embodied navigation research.

Findings

Identification of diverse task definitions and evaluation protocols

Proposal of standardized evaluation measures and scenarios

Recommendations to improve comparability and progress in navigation research

Abstract

Skillful mobile operation in three-dimensional environments is a primary topic of study in Artificial Intelligence. The past two years have seen a surge of creative work on navigation. This creative output has produced a plethora of sometimes incompatible task definitions and evaluation protocols. To coordinate ongoing and future research in this area, we have convened a working group to study empirical methodology in navigation research. The present document summarizes the consensus recommendations of this working group. We discuss different problem statements and the role of generalization, present evaluation measures, and provide standard scenarios that can be used for benchmarking.