Reproduce “LABIT: Long-Horizon Robotic Assembly Benchmark for Industrial Tasks” 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
Constantin Schempp, Christian Friedrich, Luca Kofler, Laura Suhm
LABIT benchmarks ordered robotic assembly of an e-bike-drive-inspired product, with 59 primitives across seven operation types and three subassemblies. It scores operations using estimated complexity and measured reliability, with subassembly reliability calculated as the product of primitive reliabilities. A fixed, manually engineered controller achieved 82.3% mean insertion reliability, while one reorientation-and-alignment primitive had 0% reliability; these are not end-to-end results.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.