Verifiably Following Complex Robot Instructions with Foundation Models

TL;DR

This paper introduces LIMP, a novel method enabling robots to follow complex, open-ended instructions verifiably in real-world environments without prebuilt maps, demonstrating high success rates and versatility.

Contribution

LIMP is the first approach to construct symbolic instruction representations for verifiable robot instruction following in unstructured environments.

Findings

LIMP achieves a 79% success rate on complex instructions.

LIMP performs comparably to state-of-the-art on standard tasks.

LIMP demonstrates versatility across diverse environments.

Abstract

When instructing robots, users want to flexibly express constraints, refer to arbitrary landmarks, and verify robot behavior, while robots must disambiguate instructions into specifications and ground instruction referents in the real world. To address this problem, we propose Language Instruction grounding for Motion Planning (LIMP), an approach that enables robots to verifiably follow complex, open-ended instructions in real-world environments without prebuilt semantic maps. LIMP constructs a symbolic instruction representation that reveals the robot's alignment with an instructor's intended motives and affords the synthesis of correct-by-construction robot behaviors. We conduct a large-scale evaluation of LIMP on 150 instructions across five real-world environments, demonstrating its versatility and ease of deployment in diverse, unstructured domains. LIMP performs comparably to state-of-the-art baselines on standard open-vocabulary tasks and additionally achieves a 79\% success rate on complex spatiotemporal instructions, significantly outperforming baselines that only reach 38\%. See supplementary materials and demo videos at https://robotlimp.github.io