LLM-Grounded Dynamic Task Planning with Hierarchical Temporal Logic for Human-Aware Multi-Robot Collaboration

TL;DR

This paper introduces a neuro-symbolic framework that combines Large Language Models and hierarchical Linear Temporal Logic to enable dynamic, human-aware multi-robot task planning that adapts to environmental changes in real-time.

Contribution

It presents a novel neuro-symbolic approach integrating LLM reasoning with hierarchical LTL for real-time, dynamic multi-robot task planning in human-aware environments.

Findings

Outperforms baseline methods in success rate

Reduces planning latency significantly

Enhances interaction fluency in multi-robot tasks

Abstract

While Large Language Models (LLM) enable non-experts to specify open-world multi-robot tasks, the generated plans often lack kinematic feasibility and are not efficient, especially in long-horizon scenarios. Formal methods like Linear Temporal Logic (LTL) offer correctness and optimal guarantees, but are typically confined to static, offline settings and struggle with computational scalability. To bridge this gap, we propose a neuro-symbolic framework that grounds LLM reasoning into hierarchical LTL specifications and solves the corresponding Simultaneous Task Allocation and Planning (STAP) problem. Unlike static approaches, our system resolves stochastic environmental changes, such as moving users or updated instructions via a receding horizon planning (RHP) loop with real-time perception, which dynamically refines plans through a hierarchical state space. Extensive real-world experiments demonstrate that our approach significantly outperforms baseline methods in success rate and interaction fluency while minimizing planning latency.