Driving · 2026-10-14

LIBERO-Industry: Benchmarking Robotic Manipulation under Industrial Process Constraints

Ming Shin Cindy Sung, Duan Leijiahao, Haoxiang Wang, Boning Tan, Liangna Fu, Chao Tan, Yafu Li, Yulong Chen, Chaoqun Wan, Zhen Huang, Shuxian Liang, Xian-Sheng Hua

arXiv:2610.16782PDF

AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.

TL;DR

LIBERO-Industry is a benchmark built on LIBERO to evaluate generalist robot policies under industrial process constraints by translating them into controllable conditions and measurable success criteria. It includes three task suites (S2D loading, D2S unloading, D2D transfer) totaling 216 tasks with variations in object geometry, conveyor speed, interaction-window duration, and geometric acceptance region size. Results show performance degrades as constraints accumulate and combine.

Why it matters

Industrial manipulation needs to handle product variants, moving material flows, finite execution windows, and tight geometric tolerances. The authors report that existing evaluation frameworks do not systematically capture these constraints for generalist robot policies, and they propose a benchmark that operationalizes these factors as controllable task conditions with success metrics.

Method

  • Builds LIBERO-Industry from LIBERO by defining three transfer suites (S2D, D2S, D2D) and translating industrial process constraints into controllable task conditions and measurable completion predicates.
  • Uses a language-conditioned observation-action interface; episodes use predeclared completion predicates and a task-specific rollout horizon derived from a transport schedule.
  • Evaluates five publicly available LIBERO-finetuned policies under a shared fine-tuning and evaluation protocol, and runs physical D2S trials to test sim–real co-training effects.

Limitation

The progression is described as diagnostic rather than causal because adjacent stress stages change multiple factors at once, so their differences localize breakdowns without isolating causes.

Abstract (from arXiv)

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. Such settings impose industrial process constraints, including product variants, moving material flows, finite execution windows, and tight geometric tolerances. However, existing evaluation frameworks do not systematically capture these constraints when assessing generalist robot policies. To address this gap, we introduce LIBERO-Industry, a novel benchmark built on the widely used robot learning benchmark LIBERO, that translates the aforementioned industrial constraints into controllable task conditions and measurable success criteria. Specifically, LIBERO-Industry comprises three classic task scenarios: Static-to-Dynamic (S2D) loading, Dynamic-to-Static (D2S) unloading, and Dynamic-to-Dynamic (D2D) transfer. The suites contain 216 tasks in total that systematically vary object geometry, conveyor speed, interaction-window duration, and geometric acceptance region size. Extensive experiments across five generalist robot policies (including vision-language-action models and world-action models) show that performance progressively degrades as industrial constraints are introduced and combined. Under a shared fine-tuning and evaluation protocol, the LIBERO-Industry co-training condition achieved higher observed target-transfer completion than the original-LIBERO condition.

Related papers

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First public release: 2026-10-12 · Indexed by Pub-AI: 2026-10-14 UTC

preprint

LIBERO-Industry: Benchmarking Robotic Manipulation under Industrial Process Constraints

Ming Shin Cindy Sung, Duan Leijiahao, Haoxiang Wang, Boning Tan, Liangna Fu, Chao Tan, Yafu Li, Yulong Chen, Chaoqun Wan, Zhen Huang, Shuxian Liang, Xian-Sheng Hua

Preprint posted: 2026-10-12 · Original record ↗

arxiv
2610.16782
Explicit source versions (1)
  • v2 · 2026-10-15
  1. Explicit source version detected · v2

    Source date: 2026-10-15 · First detected: 2026-10-25 UTC · Historical baseline

    Source evidence (1)
  2. Preprint available

    Source date: 2026-10-12 · First detected: 2026-10-25 UTC · Historical baseline

    Source evidence (1)
Abstract · arxiv

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. Such settings impose industrial process constraints, including product variants, moving material flows, finite execution windows, and tight geometric tolerances. However, existing evaluation frameworks do not systematically capture these constraints when assessing generalist robot policies. To address this gap, we introduce LIBERO-Industry, a novel benchmark built on the widely used robot learning benchmark LIBERO, that translates the aforementioned industrial constraints into controllable task conditions and measurable success criteria. Specifically, LIBERO-Industry comprises three classic task scenarios: Static-to-Dynamic (S2D) loading, Dynamic-to-Static (D2S) unloading, and Dynamic-to-Dynamic (D2D) transfer. The suites contain 216 tasks in total that systematically vary object geometry, conveyor speed, interaction-window duration, and geometric acceptance region size. Extensive experiments across five generalist robot policies (including vision-language-action models and world-action models) show that performance progressively degrades as industrial constraints are introduced and combined. Under a shared fine-tuning and evaluation protocol, the LIBERO-Industry co-training condition achieved higher observed target-transfer completion than the original-LIBERO condition.