2021/07/28 by Andrea Scorsoglio, Andrea D’Ambrosio, Luca Ghilardi +3 · 69 citations
Engineering · Physics and Astronomy · #Aerospace engineering #Artificial intelligence #Artificial neural network #Astro and Planetary Science #Astrobiology #Computer science #Descent (aeronautics) #Engineering #Exploration of Mars #Gradient descent #Mars Exploration Program #Mars landing #Reinforcement learning #Simulation #Space Satellite Systems and Control #Spacecraft #Spacecraft Dynamics and Control #Task (project management) #Thrust
paper · open access · doi:10.2514/1.a35072
published in Journal of Spacecraft and Rockets 59(1), 153-165 (American Institute of Aeronautics and Astronautics)
openalex publication_date 2021/07/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Future exploration and human missions on large planetary bodies (e.g., moon, Mars) will require advanced guidance navigation and control algorithms for the powered descent phase, which must be capable of unprecedented levels of autonomy. The advent of machine learning, and specifically reinforcement learning, has enabled new possibilities for closed-loop autonomous guidance and navigation. In this paper, image-based reinforcement meta-learning is applied to solve the lunar pinpoint powered descent and landing task with uncertain dynamic parameters and actuator failure. The agent, a deep neural network, takes real-time images and ranging observations acquired during the descent and maps them directly to thrust command (i.e., sensor-to-action policy). Training and validation of the algorithm and Monte Carlo simulations shows that the resulting closed-loop guidance policy reaches errors in the order of meters in different scenarios, even when the environment is partially observed, and the state of the spacecraft is not fully known.