Reproduce “From Depth to Safe Depth: Uncertainty-Aware Monocular Perception for Mobile Robot 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
Saeid Firouzi Daghigh, Saeed Ayat
The authors present an uncertainty-aware correction to monocular depth for indoor robot navigation, designed to work with an otherwise unchanged Nav2 stack. A learned head corrects predicted depths and uses uncertainty to shorten distances and handle floor pixels. In 120 paired simulated episodes, success rose from 37% to 70% and collisions fell from 63% to 3%; real-robot safety remains untested.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.