Reproduce “Don't Risk It!: Trajectory-Conditioned Distributional Risk for LiDAR-Only Crowd Navigation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Mathai Mathew Pulicken, Laksh Nanwani, Basant Sharma, Phani Teja Singamaneni, Arun Kumar Singh, K. Madhava Krishna
DRIP tackles crowd navigation by forecasting occupied space from raw two-dimensional LiDAR grids and scoring routes by uncertainty in their closest obstacle encounter. It avoids object detection and pedestrian tracking. In a simulated Lobby with 45 pedestrians and 0.05-meter Gaussian LiDAR noise, DRIP achieved 86% success and a 13.4-second completion time, versus SCOPE’s 64% success and 27.0 seconds. Success required reaching the goal without collision or timeout.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.