Constrained Robotic Navigation on Preferred Terrains Using LLMs and Speech Instruction: Exploiting the Power of Adverbs

TL;DR

This paper presents a novel approach to off-road robotic navigation that uses large language models and speech instructions to interpret high-level commands, identify landmarks and terrains, and control vehicle movement without relying on extensive data collection.

Contribution

The paper introduces a method combining LLMs, speech-to-text, semantic segmentation, and MPC control for map-free, terrain-preferred navigation based on verbal instructions and adverbs.

Findings

Effective interpretation of verbal instructions for navigation.

Successful identification of landmarks and terrains from images.

Enhanced adaptability to diverse off-road environments.

Abstract

This paper explores leveraging large language models for map-free off-road navigation using generative AI, reducing the need for traditional data collection and annotation. We propose a method where a robot receives verbal instructions, converted to text through Whisper, and a large language model (LLM) model extracts landmarks, preferred terrains, and crucial adverbs translated into speed settings for constrained navigation. A language-driven semantic segmentation model generates text-based masks for identifying landmarks and terrain types in images. By translating 2D image points to the vehicle's motion plane using camera parameters, an MPC controller can guides the vehicle towards the desired terrain. This approach enhances adaptation to diverse environments and facilitates the use of high-level instructions for navigating complex and challenging terrains.