If we want to train robots in simulation before deploying them in reality, it seems natural and almost self-evident to presume that reducing the sim2real gap involves creating simulators of increasing fidelity (since reality is what it is). We challenge this assumption and present a contrary hypothesis -- sim2real transfer of robots may be improved with lower (not higher) fidelity simulation. We conduct a systematic large-scale evaluation of this hypothesis on the problem of visual navigation -- in the real world, and on 2 different simulators (Habitat and iGibson) using 3 different robots (A1, AlienGo, Spot). Our results show that, contrary to expectation, adding fidelity does not help with learning; performance is poor due to slow simulation speed (preventing large-scale learning) and overfitting to inaccuracies in simulation physics. Instead, building simple models of the robot motion using real-world data can improve learning and generalization.
@article{arxiv.2207.10821,
title = {Rethinking Sim2Real: Lower Fidelity Simulation Leads to Higher Sim2Real Transfer in Navigation},
author = {Joanne Truong and Max Rudolph and Naoki Yokoyama and Sonia Chernova and Dhruv Batra and Akshara Rai},
journal= {arXiv preprint arXiv:2207.10821},
year = {2022}
}