LIBERO-Industry benchmarks robot policies under factory process constraints
Original title: LIBERO-Industry: Benchmarking Robotic Manipulation under Industrial Process Constraints
Five generalist policies, including vision-language-action models, degrade as industrial constraints are combined.
LIBERO-Industry extends the LIBERO robot-learning benchmark to high-mix, low-volume manufacturing constraints such as product variants, moving material flows, finite windows, and tight tolerances. It defines Static-to-Dynamic loading, Dynamic-to-Static unloading, and Dynamic-to-Dynamic transfer. Across 216 tasks, five generalist robot policies lose performance as constraints are added. Under a shared protocol, co-training on LIBERO-Industry beat the original-LIBERO condition on observed target-transfer completion.
- Tasks vary object geometry, conveyor speed, window duration, and acceptance-region size.
- Policies tested include vision-language-action and world-action models.
- The suites contain 216 tasks in total.
- Co-training raised observed target-transfer completion versus original LIBERO.
From the original article
1 min read at huggingface.co
Industrial manipulation in flexible manufacturing, e. g. , high-mix, low-volume (HMLV) electronics assembly, requires robot policies to adapt to changing products and production conditions.
By Ming Shin Cindy Sung, Duan Leijiahao, Haoxiang Wang · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.