Search papers, labs, and topics across Lattice.
AeroGen is introduced, an open-loop framework leveraging structured guardrail prompting and the AeroDaaS drone SDK to generate correct single-shot AI-generated drone control programs. By encoding API descriptions, flight constraints, and operational world rules directly into the system context prompt, AeroGen enables generic LLMs to produce constraint-aware code from user prompts with minimal examples. Evaluations across 20 navigation tasks and 5 drone missions demonstrate that AeroGen generates approximately 40 lines of AeroDaaS Python code in about 20 seconds per mission, improving the robustness, correctness, and deployability of LLM-generated drone autonomy programs.
Forget hand-coding drone missions – AeroGen uses structured prompts to let LLMs reliably generate safe, constraint-aware flight code in seconds.
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.
