Egocentric Tactile and Proximity Sensors as Observation Priors for Humanoid Collision Avoidance
For humanoid robotics, this work provides empirical guidance on sensor design for collision avoidance, though the findings are specific to the dodgeball task and may not generalize broadly.
This work presents a reinforcement learning framework for whole-body collision avoidance on a humanoid H1-2 robot, using dodgeball as a benchmark to characterize how sensor properties (coverage, type, range) shape learned avoidance behavior. Key findings include that raw proximity measurements can substitute for explicit object localization given sufficient range, and sparse non-directional proximity signals outperform dense directional alternatives in sample efficiency.
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.