持续语言引导的机器人操作学习
Lifelong Language-Conditioned Robotic Manipulation Learning
AI总结：
本文提出SkillsCrafter框架，通过操作技能适应和技能专业化聚合，实现持续学习多种技能并减少旧技能的灾难性遗忘。
摘要：
传统语言引导的操作代理在连续适应新操作技能时会导致旧技能的灾难性遗忘，限制了动态场景中的实际应用。在本文中，我们提出SkillsCrafter，一种新的机器人操作框架，旨在持续学习多种技能的同时减少旧技能的灾难性遗忘。具体来说，我们提出操作技能适应，以保留旧技能的知识，同时继承新旧技能之间的共享知识，以促进新技能的学习。同时，我们对多样的技能指令进行奇异值分解，以获得共同的技能语义子空间投影矩阵，从而记录技能的本质语义空间。为了实现无遗忘和泛化操作，我们提出技能专业化聚合，以在技能语义子空间中计算技能间的相似性，从而为任何新的或未知的技能聚合先前学习的技能知识。广泛的实验展示了我们提出的SkillsCrafter的有效性和优越性。
英文摘要：
Traditional language-conditioned manipulation agent sequential adaptation to new manipulation skills leads to catastrophic forgetting of old skills, limiting dynamic scene practical deployment. In this paper, we propose SkillsCrafter, a novel robotic manipulation framework designed to continually learn multiple skills while reducing catastrophic forgetting of old skills. Specifically, we propose a Manipulation Skills Adaptation to retain the old skills knowledge while inheriting the shared knowledge between new and old skills to facilitate learning of new skills. Meanwhile, we perform the singular value decomposition on the diverse skill instructions to obtain common skill semantic subspace projection matrices, thereby recording the essential semantic space of skills. To achieve forget-less and generalization manipulation, we propose a Skills Specialization Aggregation to compute inter-skills similarity in skill semantic subspaces, achieving aggregation of the previously learned skill knowledge for any new or unknown skill. Extensive experiments demonstrate the effectiveness and superiority of our proposed SkillsCrafter.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
