Learning and Autonomy for Space Exploration: An Experience Report from OWLAT Deployment

Extraterrestrial autonomous lander missions are increasingly demanding adaptive capabilities to handle the unpredictable and diverse nature of the terrain. This paper discusses the deployment of a Deep Meta-Learning with Controlled Deployment Gaps (CoDeGa) trained model for terrain scooping tasks in Ocean Worlds Lander Autonomy Testbed (OWLAT) at NASA Jet Propulsion Laboratory. The CoDeGa-powered scooping strategy is designed to adapt to novel terrains, selecting scooping actions based on the available RGB-D image data and limited experience. The paper presents our experiences with transferring the CoDeGa model trained on a low-fidelity UIUC testbed to the high-fidelity OWLAT evaluating the feasibility of transferring the model to similar systems, validating the model’s performance in novel, realistic environments, and sharing the lessons learned from deploying learning-based autonomy algorithms for space exploration. Preliminary findings substantiate the efficacy of CoDeGA in adapting to unfamiliar terrains, effectively making autonomous decisions under considerable domain shifts, thereby endorsing its potential utility in future extraterrestrial missions.

地外自主着陆器任务愈发需要自适应能力，以应对地形不可预测且多样的特性。本文探讨了在美国国家航空航天局（NASA）喷气推进实验室的海洋世界着陆器自主测试平台（Ocean Worlds Lander Autonomy Testbed, OWLAT）中，部署经带受控部署间隙的深度元学习（Deep Meta-Learning with Controlled Deployment Gaps, CoDeGa）训练得到的模型以完成地形铲取任务的相关工作。该基于CoDeGa的铲取策略旨在适配未知地形，可基于可用的RGB-D图像数据与有限经验选择铲取动作。本文分享了将在低保真度UIUC测试平台上训练得到的CoDeGa模型迁移至高保真度OWLAT平台的实践经验，旨在评估模型向同类系统迁移的可行性，验证模型在未知且贴近真实的环境中的性能，并分享部署面向深空探测的基于学习的自主算法所得到的经验教训。初步研究结果证实了CoDeGa在适配陌生地形方面的有效性，其可在显著的域偏移下自主做出有效决策，从而证明了该方法在未来地外任务中的应用潜力。