Egocentric Tactile and Proximity Sensors as Observation Priors for Humanoid Collision Avoidance
Carson Kohlbrenner, Niraj Pudasaini, William Xie, Naren Sivagnanadasan, Nikolaus Correll + 1 more
TLDR
This paper uses an RL framework to characterize how egocentric tactile and proximity sensor properties affect humanoid collision avoidance.
Key contributions
- Developed an RL framework for whole-body collision avoidance on a H1-2 humanoid robot.
- Characterized how sensor properties (coverage, type, range) influence learned avoidance behaviors.
- Found raw proximity measurements can replace explicit object localization with sufficient range.
- Sparse, non-directional proximity signals showed higher sample efficiency than dense, directional ones.
Why it matters
This work provides crucial insights into designing effective egocentric sensor systems for humanoid robots. Understanding optimal sensor properties can significantly improve collision avoidance, enhancing robot safety and autonomy in complex environments.
Original Abstract
Collision-free motion is often aided by tactile and proximity sensors distributed on the body of the robot due to their resistance to occlusion as opposed to external cameras. However, how to shape the sensor's properties, such as sensing coverage; type; and range, to enable avoidant behavior remains unclear. In this work, we present a reinforcement learning framework for whole-body collision avoidance on a humanoid H1-2 robot and use it to characterize how sensor properties shape learned avoidance behavior. Using dodgeball as a benchmark task, we ablate the properties of sensors distributed across the upper body of the robot and find that raw proximity measurements can substitute for explicit object localization provided the sensing range is sufficient and that sparse non-directional proximity signals outpace dense directional alternatives in sample efficiency.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.