AeroGen: Agentic Drone Autonomy through Single-Shot Structured Prompting & Drone SDK

TL;DR

AeroGen leverages structured prompting and a specialized SDK to enable reliable, constraint-aware drone control programs generated by LLMs, improving safety and robustness across diverse navigation tasks.

Contribution

The paper introduces AeroGen, a novel framework that integrates structured guardrail prompts with a drone SDK to produce correct, constraint-aware drone control code from LLMs.

Findings

AeroGen generates about 40 lines of code in 20 seconds per mission.

Structured prompting improves robustness and correctness of LLM-generated drone programs.

AeroGen performs well across urban, farm, and inspection environments.

Abstract

Designing correct UAV autonomy programs is challenging due to joint navigation, sensing and analytics requirements. While LLMs can generate code, their reliability for safety-critical UAVs remains uncertain. This paper presents AeroGen, an open-loop framework that enables consistently correct single-shot AI-generated drone control programs through structured guardrail prompting and integration with the AeroDaaS drone SDK. AeroGen encodes API descriptions, flight constraints and operational world rules directly into the system context prompt, enabling generic LLMs to produce constraint-aware code from user prompts, with minimal example code. We evaluate AeroGen across a diverse benchmark of 20 navigation tasks and 5 drone missions on urban, farm and inspection environments, using both imperative and declarative user prompts. AeroGen generates about 40 lines of AeroDaaS Python code in about 20s per mission, in both real-world and simulations, showing that structured prompting with a well-defined SDK improves robustness, correctness and deployability of LLM-generated drone autonomy programs.