PulseAugur
LIVE 12:26:07
research · [2 sources] ·
0
research

Humanoid robots use egocentric sensors for collision avoidance

Researchers have developed a reinforcement learning framework to improve collision avoidance in humanoid robots. The study focused on how the properties of egocentric tactile and proximity sensors, such as their range and coverage, influence a robot's ability to avoid obstacles. Experiments using a dodgeball task on the H1-2 robot demonstrated that raw proximity data can be effective for avoidance if the sensing range is adequate, and that sparse, non-directional sensors can be more sample-efficient than dense, directional ones. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT Enhances robot safety and efficiency in dynamic environments, potentially improving humanoid robot performance in complex tasks.

RANK_REASON Academic paper detailing a new reinforcement learning framework for robot collision avoidance.

Read on arXiv cs.LG →

COVERAGE [2]

  1. arXiv cs.LG TIER_1 · Carson Kohlbrenner, Niraj Pudasaini, William Xie, Naren Sivagnanadasan, Nikolaus Correll, Alessandro Roncone ·

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

    arXiv:2604.25554v1 Announce Type: cross 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…

  2. arXiv cs.LG TIER_1 · Alessandro Roncone ·

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

    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 avoi…