No RL, No Simulation: Learning to Navigate without NavigatingDownload PDF

21 May 2021, 20:45 (modified: 21 Jan 2022, 16:37)NeurIPS 2021 PosterReaders: Everyone
Keywords: Navigation, Simulation, Graph Neural Networks, Embodied AI
TL;DR: We present a navigation method which does not rely on RL or simulation. It uses simple self-supervised distance learning from passive house tour videos along with greedy decision-making to outperform RL/IL algorithms on image-goal navigation.
Abstract: Most prior methods for learning navigation policies require access to simulation environments, as they need online policy interaction and rely on ground-truth maps for rewards. However, building simulators is expensive (requires manual effort for each and every scene) and creates challenges in transferring learned policies to robotic platforms in the real-world, due to the sim-to-real domain gap. In this paper, we pose a simple question: Do we really need active interaction, ground-truth maps or even reinforcement-learning (RL) in order to solve the image-goal navigation task? We propose a self-supervised approach to learn to navigate from only passive videos of roaming. Our approach, No RL, No Simulator (NRNS), is simple and scalable, yet highly effective. NRNS outperforms RL-based formulations by a significant margin. We present NRNS as a strong baseline for any future image-based navigation tasks that use RL or Simulation.
Supplementary Material: pdf
Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.
18 Replies