COA-GPT: Generative Pre-Trained Transformers for Accelerated Course of Action Development in Military Operations

The development of Courses of Action (COAs) in military operations is traditionally a time-consuming and intricate process. Addressing this challenge, this study intro-duces COA-GPT, a novel algorithm employing Large Language Models (LLMs) for rapid and efficient generation of valid COAs. COA-GPT incorporates military doctrine excerpts and domain expertise to LLMs through in-context learning, allowing commanders to input mission information - in both text and image formats - and receive strategically aligned COAs for review and approval. Uniquely, COA-GPT not only accelerates COA development, producing initial COAs within seconds, but also facilitates real-time refinement based on commander feedback. This work evaluates COA-GPT in a military-relevant scenario within a militarized version of the StarCraft II game, comparing its performance against an expert human and state-of-the-art re-inforcement learning algorithms. Our results demonstrate COA-GPT's superiority in generating strategically sound COAs more swiftly, with the added benefits of enhanced adaptability and alignment with commander intentions. COA-GPT's capability to rapidly adapt and update COAs during missions presents a transformative potential for military planning, particularly in addressing planning discrepancies and capitalizing on emergent windows of opportunity. Performance videos of our method can be seen at https://sites.google.com/view/coa-gpt.