arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.09080cs.ROcs.LG
收藏

约束流形上的安全强化学习:理论与应用

Safe Reinforcement Learning on the Constraint Manifold: Theory and Applications

Technical University of Darmstadt(达姆施塔特工业大学);Huawei R&D London(华为伦敦研发中心);German Research Center for AI(德国人工智能研究中心);Hessian Centre for Artificial Intelligence(黑森州人工智能中心);Centre of Cognitive Science(认知科学中心)

Puze Liu, Haitham Bou-Ammar, Jan Peters, Davide Tateo

AI总结:

本文提出基于约束流形与切空间构建安全动作空间的方法,在保证安全的前提下允许智能体任意采样动作,并在真实机器人冰球任务中验证了其处理高维复杂约束的有效性。

摘要:

将基于学习的技术,特别是强化学习,集成到机器人技术中,有望解决非结构化环境中的复杂问题。然而,现有的大多数方法都是在精心调整的模拟器中训练,随后部署到真实机器人上,而不进行在线微调。在这种设置下,需要大量的工程来缓解模拟到现实的差距,这对于复杂系统可能具有挑战性。相反,使用真实世界交互数据进行学习提供了一种有前景的替代方案:它不仅消除了对精细模拟器的需求,而且适用于更广泛的任务,在这些任务中,精确建模是不可行的。机器人上强化学习的一个主要问题是确保安全性,因为不受控制的探索可能对机器人或环境造成灾难性损害。实际上,安全规范通常表示为约束,可能是复杂且非线性的,使得在学习系统中保证安全性具有挑战性。在本文中,我们从理论和实践的角度展示了如何以原则性的方式将复杂的安全约束施加于基于学习的机器人系统。我们的方法基于约束流形的概念,它表示安全机器人配置的集合。利用微分几何技术,即切空间,我们可以构建一个安全动作空间,允许学习代理采样任意动作,同时确保安全性。我们在真实世界的机器人冰球任务中展示了该方法的有效性,表明我们的方法可以处理具有复杂约束的高维任务。真实机器人实验的视频可在项目网站上获取(https://puzeliu.github.io/TRO-ATACOM)。

英文摘要:

Integrating learning-based techniques, especially reinforcement learning, into robotics is promising for solving complex problems in unstructured environments. However, most existing approaches are trained in well-tuned simulators and subsequently deployed on real robots without online fine-tuning. In this setting, extensive engineering is required to mitigate the sim-to-real gap, which can be challenging for complex systems. Instead, learning with real-world interaction data offers a promising alternative: it not only eliminates the need for a fine-tuned simulator but also applies to a broader range of tasks where accurate modeling is unfeasible. One major problem for on-robot reinforcement learning is ensuring safety, as uncontrolled exploration can cause catastrophic damage to the robot or the environment. Indeed, safety specifications, often represented as constraints, can be complex and non-linear, making safety challenging to guarantee in learning systems. In this paper, we show how we can impose complex safety constraints on learning-based robotics systems in a principled manner, both from theoretical and practical points of view. Our approach is based on the concept of the Constraint Manifold, representing the set of safe robot configurations. Exploiting differential geometry techniques, i.e., the tangent space, we can construct a safe action space, allowing learning agents to sample arbitrary actions while ensuring safety. We demonstrate the method's effectiveness in a real-world Robot Air Hockey task, showing that our method can handle high-dimensional tasks with complex constraints. Videos of the real robot experiments are available on the project website (https://puzeliu.github.io/TRO-ATACOM).

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。

↑