aiqbench analysis
This paper studies how the design of a robot's sensing inputs affects how quickly a reinforcement learning policy learns to dodge collisions. Its most useful finding is that coarse, lossy proximity sensing can serve as a better inductive bias than dense, high-resolution coverage, and that explicitly predicting contact can slow learning. Results come from simulated humanoid dodgeball across eight representations, so transfer to physical robots remains an open question.
Move, don't think: sample-efficient collision avoidance with lossy proximity sensors
arXiv:2610.31891v1 Announce Type: new Abstract: Data-driven robots must learn to avoid collisions with as few demonstrations as possible to avoid collecting large datasets of destructive collision examples. We explore whole-body proximity sensors as inductive biases for learning reflexive collision-dodging behavior and find that lossy sensors with low resolution were stronger biases than high-resolution coverage. We also found that, perhaps unintuitively, anticipating when and where contact happens on the body of a robot reduced its learning sample efficiency. We investigated eight candidate representations in a reinforcement learning environment, which included latent compression, future contact anticipation, lossy corruption, and privileged state estimation. The candidate representations were tested by training over 200 humanoid robots to play a simplified version of dodgeball end-to-end under an intentionally constrained sample and compute budget.
Abstract:Data-driven robots must learn to avoid collisions with as few demonstrations as possible to avoid collecting large datasets of destructive collision examples. We explore whole-body proximity sensors as inductive biases for learning reflexive collision-dodging behavior and find that lossy sensors with low resolution were stronger biases than high-resolution coverage. We also found that, perhaps unintuitively, anticipating when and where contact happens on the body of a robot reduced its learning sample efficiency. We investigated eight candidate representations in a reinforcement learning environment, which included latent compression, future contact anticipation, lossy corruption, and privileged state estimation. The candidate representations were tested by training over 200 humanoid robots to play a simplified version of dodgeball end-to-end under an intentionally constrained sample and compute budget.
Robotics (cs.RO)
arXiv:2610.31891 [cs.RO]
(or arXiv:2610.31891v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.31891 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 215 words