Reproduce “A Sim-to-Real Deep Learning-based Framework for Autonomous Nano-drone Racing” 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
Lorenzo Lamberti, Elia Cereda, Gabriele Abbate, Lorenzo Bellone, Victor Javier Kartsch Morinigo, Michał Barcis, Agata Barcis, Alessandro Giusti and 2 more
The authors train a visual obstacle-avoidance network on simulated images and pair it with waypoint guidance based on known field boundaries for a nano-drone. In five-minute arena tests, Policy 2’s median distance was 91 m, versus 45 m for Policy 1 at 1.5 m/s. At the 2022 challenge, the drone covered 115 m without crashing, but passed through no gates.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.