Egocentric Tactile and Proximity Sensors as Observation Priors for Humanoid Collision Avoidance

TL;DR

This paper introduces a reinforcement learning framework for humanoid collision avoidance using tactile and proximity sensors, analyzing how sensor properties influence avoidance behavior.

Contribution

It characterizes the impact of sensor properties on learned avoidance behavior and demonstrates the effectiveness of raw proximity measurements in a benchmark task.

Findings

Raw proximity measurements can replace explicit object localization with sufficient sensing range.

Sparse non-directional proximity signals outperform dense directional signals in sample efficiency.

Sensor property ablation reveals key factors for effective collision avoidance.

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.