Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LLM-Enabled Low-Altitude UAV Natural Language Navigation via Signal Temporal Logic Specification Translation and Repair

Yuqi Ping, Huahao Ding, Tianhao Liang, Longyu Zhou, Guangyu Lei, Xinglin Chen, Junwei Wu, Jieyu Zhou, Tingting Zhang

Abstract

Natural language (NL) navigation for low-altitude unmanned aerial vehicles (UAVs) offers an intelligent and convenient solution for low-altitude aerial services by enabling an intuitive interface for non-expert operators. However, deploying this capability in urban environments necessitates the precise grounding of underspecified instructions into safety-critical, dynamically feasible motion plans subject to spatiotemporal constraints. To address this challenge, we propose a unified framework that translates NL instructions into Signal Temporal Logic (STL) specifications and subsequently synthesizes trajectories via mixed-integer linear programming (MILP). Specifically, to generate executable STL formulas from free-form NL, we develop a reasoning-enhanced large language model (LLM) leveraging chain-of-thought (CoT) supervision and group-relative policy optimization (GRPO), which ensures high syntactic validity and semantic consistency. Furthermore, to resolve infeasibilities induced by stringent logical or spatial requirements, we introduce a specification repair mechanism. This module combines MILP-based diagnosis with LLM-guided semantic reasoning to selectively relax task constraints while strictly enforcing safety guarantees. Extensive simulations and real-world flight experiments demonstrate that the proposed closed-loop framework significantly improves NL-to-STL translation robustness, enabling safe, interpretable, and adaptable UAV navigation in complex scenarios.

LLM-Enabled Low-Altitude UAV Natural Language Navigation via Signal Temporal Logic Specification Translation and Repair

Abstract

Natural language (NL) navigation for low-altitude unmanned aerial vehicles (UAVs) offers an intelligent and convenient solution for low-altitude aerial services by enabling an intuitive interface for non-expert operators. However, deploying this capability in urban environments necessitates the precise grounding of underspecified instructions into safety-critical, dynamically feasible motion plans subject to spatiotemporal constraints. To address this challenge, we propose a unified framework that translates NL instructions into Signal Temporal Logic (STL) specifications and subsequently synthesizes trajectories via mixed-integer linear programming (MILP). Specifically, to generate executable STL formulas from free-form NL, we develop a reasoning-enhanced large language model (LLM) leveraging chain-of-thought (CoT) supervision and group-relative policy optimization (GRPO), which ensures high syntactic validity and semantic consistency. Furthermore, to resolve infeasibilities induced by stringent logical or spatial requirements, we introduce a specification repair mechanism. This module combines MILP-based diagnosis with LLM-guided semantic reasoning to selectively relax task constraints while strictly enforcing safety guarantees. Extensive simulations and real-world flight experiments demonstrate that the proposed closed-loop framework significantly improves NL-to-STL translation robustness, enabling safe, interpretable, and adaptable UAV navigation in complex scenarios.

Paper Structure

This paper contains 28 sections, 32 equations, 9 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Motivation
  3. Related Works
  4. Main Contributions
  5. Preliminaries and Problem Formulation
  6. UAV Dynamics and Environment Model
  7. Signal Temporal Logic Specifications
  8. Problem Definition
  9. System Overview and Framework
  10. Reasoning-Enhanced NL to STL Translation
  11. Synthetic Dataset and CoT Data Generation
  12. Cold Start Stage
  13. GRPO Reinforcement Learning
  14. STL-Constrained Trajectory Planning and Specification Repair
  15. MILP Encoding of STL Satisfaction
  16. ...and 13 more sections

Figures (9)

  • Figure 1: Overview of the proposed language-guided planning framework. A natural-language instruction is translated into an STL specification, which is enforced by a constrained planner. When infeasibility is detected, solver feedback is used to guide specification refinement and re-planning in a closed loop.
  • Figure 2: GRPO-based RL framework for reasoning-enhanced NL-to-STL generation.
  • Figure 3: CoT Data Engine for augmenting NL2TL with intermediate reasoning traces.
  • Figure 4: NL-to-STL translation accuracy during SFT and GRPO training.
  • Figure 5: Simulated trajectory for Task 1. The UAV reaches the goal region within the specified time window while avoiding obstacles.
  • ...and 4 more figures