On Evaluation of Embodied Navigation Agents

TLDR

Artificial Intelligence research focuses on skillful mobile operation in 3‑D environments, and recent work has produced many diverse and sometimes incompatible navigation tasks and evaluation protocols. The working group was convened to coordinate research and produce consensus recommendations on empirical methodology, problem statements, generalization, evaluation measures, and benchmark scenarios for navigation. The authors convened a working group that examined empirical methodology, problem statements, generalization, evaluation metrics, and benchmark scenarios for 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.