{"slug": "an-rl-agent-learning-to-land-a-falcon-9-on-a-droneship-in-simulation", "title": "An RL agent learning to land a Falcon 9 on a droneship in simulation", "summary": "An independent developer trained a reinforcement-learning agent to land a simulated Falcon 9 booster on a droneship, using only trial and error with no real flight data. The simulation models a hoverslam, in which the booster cannot hover and must reach zero speed exactly at the deck.", "body_md": "I trained an AI with reinforcement learning to land a Falcon 9 on a droneship, in a fully simulated world.\nNo real flight data, just trial and error. \nAnd it’s a hoverslam: the booster can’t hover, so it has to hit zero speed exactly at the deck.\n\n00:00", "url": "https://wpnews.pro/news/an-rl-agent-learning-to-land-a-falcon-9-on-a-droneship-in-simulation", "canonical_source": "https://twitter.com/breeg554/status/2112484802222072189", "published_at": "2026-10-20 10:13:27+00:00", "updated_at": "2026-10-20 10:47:11.953195+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-research"], "entities": ["Falcon 9", "SpaceX"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/an-rl-agent-learning-to-land-a-falcon-9-on-a-droneship-in-simulation", "markdown": "https://wpnews.pro/news/an-rl-agent-learning-to-land-a-falcon-9-on-a-droneship-in-simulation.md", "text": "https://wpnews.pro/news/an-rl-agent-learning-to-land-a-falcon-9-on-a-droneship-in-simulation.txt", "jsonld": "https://wpnews.pro/news/an-rl-agent-learning-to-land-a-falcon-9-on-a-droneship-in-simulation.jsonld"}}