Constrained Robotic Navigation on Preferred Terrains Using LLMs and Speech Instruction: Exploiting the Power of Adverbs
This work addresses the challenge of adapting robots to diverse and complex terrains using natural language instructions, though it appears incremental as it combines existing components like LLMs and MPC controllers.
The paper tackles the problem of map-free off-road robotic navigation by developing a system that converts verbal instructions into constrained navigation commands using large language models and semantic segmentation, enabling robots to follow high-level terrain preferences without traditional data collection.
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.