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

2026-03-15

AI Summary

This work proposes AeroGen, an open-loop code generation framework that enables general-purpose large language models (LLMs) to produce safe, deployable drone control code without requiring examples or fine-tuning. Addressing the challenge of unreliable LLM-generated code in safety-critical autonomous systems—where navigation, perception, and decision-making must be tightly integrated—AeroGen embeds flight constraints, operational rules, and API specifications from the AeroDaaS drone SDK into structured system prompts. This approach leverages structured guardrail prompting to guide the LLM toward generating correct, constraint-compliant Python scripts in a single pass. Evaluated across 20 navigation tasks and five real-world and simulated scenarios—including urban, agricultural, and inspection environments—AeroGen consistently produces approximately 40 lines of functional code within 20 seconds on average, supporting both imperative and declarative inputs while demonstrating high robustness and practical utility.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

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.

Problem

• UAV autonomy
• LLM reliability
• drone control programs
• safety-critical systems
• code generation

Innovation

• structured prompting
• drone autonomy
• LLM code generation
• AeroDaaS SDK
• safety-critical UAV