Abstract
With the rapid advancement of Large Language Models (LLMs), their capabilities in various automation domains, particularly Unmanned Aerial Vehicle (UAV) operations, have gathered increasing attention. Current research remains predominantly constrained to the deployment of LLMs in small-scale UAV applications, with most studies focusing on isolated components such as path planning for toy drones, while lacking comprehensive investigation of medium- and long-range UAV systems in real-world operational contexts. Larger UAV platforms introduce distinct challenges, including stringent requirements for airport-based take-off and landing procedures, adherence to complex regulatory frameworks, and specialized operational capabilities with elevated mission expectations. This paper presents the Next-Generation LLM for UAV (NeLV) system, a comprehensive framework that seamlessly integrates LLMs into multi-scale UAV operations. The NeLV system processes natural language instructions to orchestrate short-, medium-, and long-range UAV missions through five key technical components: (i) LLM-as-Parser for instruction interpretation, (ii) Route Planner for Points of Interest (POI) determination, (iii) Path Planner for waypoint generation, (iv) Control Platform for executable trajectory implementation, and (v) real UAV monitoring capabilities. We validate the system through three representative use cases: multi-UAV patrol, multi-POI delivery, and multi-hop relocation. This research not only synthesizes existing literature and demonstrates practical implementation but also establishes a comprehensive roadmap featuring a five-level automation taxonomy for the future development of LLM-powered autonomous aerial systems.
Use Cases
Use Case 1: Short-Range Multi-UAV Patrol
LLM
Path
Route
Traj
Use Case 2: Medium-Range Multi-POI Delivery
LLM
Path
Route
Traj
Use Case 3: Long-Range Multi-Hop Relocation
LLM
Path
Route
Traj
NeLV: Next-Generation LLM for UAV System
TL;DR: NeLV is an end-to-end framework that encompasses the entire operational pipeline from human instruction input to UAV flight mission execution.
Overview
NeLV is a versatile platform that currently supports:
- Range Features: Short-, medium-, and long-range operations
- Operational Functions: Multi-UAV, multi-POI, and multi-hop coordinations
- Mission Types: Patrol, delivery, and relocation operations
Operational Framework
The system comprises five interconnected components:
- LLM-as-Parser: Interprets natural language mission instructions
- Route Planner: Optimizes flight routes based on mission objectives
- Path Planner: Generates safe waypoints avoiding restricted zones and weather risks
- Control Platform: Converts missions and waypoints into executable flight trajectories
- UAV Monitor: Provides real-time mission oversight and intervention capabilities
Roadmap and Future Directions
Our five-level roadmap envisions the progressive evolution of LLM capabilities in UAV operations. While current LLMs demonstrate substantial potential, they require integration with specialized UAV domain knowledge including airspace regulations, aeronautical data, and control systems. The modular NeLV architecture anticipates gradual component consolidation as UAV-specific language models mature, ultimately progressing toward fully autonomous mission planning and execution systems.
Video Presentation
BibTeX
@article{yuan2025next,
title={Next-Generation LLM for UAV: From Natural Language to Autonomous Flight},
author={Yuan, Liangqi and Deng, Chuhao and Han, Dong-Jun and Hwang, Inseok and Brunswicker, Sabine and Brinton, Christopher G},
journal={To appear},
year={2025}
}