Adaptive Interactive Navigation of Quadruped Robots using Large Language Models

2025-03-29

AI Summary

Conventional quadrupedal robot navigation methods fail to reach unreachable targets in complex, pathless environments—e.g., disaster zones or cluttered warehouses—where static path planning is infeasible. Method: We propose an LLM-driven interactive navigation framework featuring a hierarchical action tree to represent environment interactions; a dual-module adaptive replanning mechanism (“Consultant Trigger–Gardener Adjustment”) enabling dynamic node insertion/deletion and real-time trajectory correction; and tight integration of LLM-based task reasoning, reinforcement-learning pre-trained skill libraries, and hierarchical action execution. Contribution/Results: Evaluated in diverse unknown complex scenarios via simulation and real-robot experiments, our approach significantly improves navigation success rate and environmental adaptability. It achieves, for the first time, autonomous physical interaction—such as obstacle pushing and rubble trampling—to dynamically construct traversable paths and reach previously unreachable goals, thereby validating the efficacy of interactive path generation as a novel navigation paradigm.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

Robotic navigation in complex environments remains a critical research challenge. Traditional navigation methods focus on optimal trajectory generation within free space, struggling in environments lacking viable paths to the goal, such as disaster zones or cluttered warehouses. To address this gap, we propose an adaptive interactive navigation approach that proactively interacts with environments to create feasible paths to reach originally unavailable goals. Specifically, we present a primitive tree for task planning with large language models (LLMs), facilitating effective reasoning to determine interaction objects and sequences. To ensure robust subtask execution, we adopt reinforcement learning to pre-train a comprehensive skill library containing versatile locomotion and interaction behaviors for motion planning. Furthermore, we introduce an adaptive replanning method featuring two LLM-based modules: an advisor serving as a flexible replanning trigger and an arborist for autonomous plan adjustment. Integrated with the tree structure, the replanning mechanism allows for convenient node addition and pruning, enabling rapid plan modification in unknown environments. Comprehensive simulations and experiments have demonstrated our method's effectiveness and adaptivity in diverse scenarios. The supplementary video is available at page: https://youtu.be/W5ttPnSap2g.

Problem

• Enables robots to navigate complex environments by interacting with obstacles
• Uses LLMs for adaptive task planning and interaction sequence reasoning
• Integrates reinforcement learning for robust locomotion and interaction behaviors

Innovation

• LLM-based primitive tree for task planning
• Reinforcement learning pre-trained skill library
• Adaptive replanning with LLM advisor and arborist