LEMMA: Learning Language-Conditioned Multi-Robot Manipulation

TL;DR

LEMMA introduces a challenging benchmark for language-conditioned multi-robot manipulation tasks, emphasizing task allocation, tool use, and long-horizon planning, with a modular hierarchical approach as a baseline.

Contribution

The paper presents a new benchmark, LEMMA, for multi-robot manipulation based on language instructions, including diverse tasks, expert demonstrations, and a baseline planning method.

Findings

LEMMA is more challenging than existing benchmarks.

The proposed hierarchical planning approach provides a baseline performance.

Results demonstrate the potential for future development of language-conditioned multi-robot systems.

Abstract

Complex manipulation tasks often require robots with complementary capabilities to collaborate. We introduce a benchmark for LanguagE-Conditioned Multi-robot MAnipulation (LEMMA) focused on task allocation and long-horizon object manipulation based on human language instructions in a tabletop setting. LEMMA features 8 types of procedurally generated tasks with varying degree of complexity, some of which require the robots to use tools and pass tools to each other. For each task, we provide 800 expert demonstrations and human instructions for training and evaluations. LEMMA poses greater challenges compared to existing benchmarks, as it requires the system to identify each manipulator's limitations and assign sub-tasks accordingly while also handling strong temporal dependencies in each task. To address these challenges, we propose a modular hierarchical planning approach as a baseline. Our results highlight the potential of LEMMA for developing future language-conditioned multi-robot systems.