arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.06045eess.SYcs.SY
收藏

连续控制下样本最优的零违反安全性

Sample-Optimal Zero-Violation Safety For Continuous Control

Ritabrata Ray, Yorie Nakahira, Soummya Kar

AI总结:

针对部分未知系统少样本安全控制问题,刻画了样本不足时的安全极限,提出了用最少样本生成可证安全动作与恢复行为的方法,还得到了宽松假设下的类Nagumo定理结果,并可集成到策略梯度算法中实现少样本安全稳定控制。

摘要:

本文研究了部分未知系统中利用少量样本确保安全性的问题。我们首先刻画了因信息或样本不足而无法生成安全动作时的基本极限。随后,我们开发了一种技术,能够使用最少数量的样本生成可证明安全的动作和恢复行为。在性能分析中,我们还在更宽松的假设下建立了类似Nagumo定理的结果,该结果在其他场景中也具备潜在应用价值。最后,我们讨论了如何将所提方法集成到策略梯度算法中,以仅用少量样本即可保证安全性和稳定性,且无需稳定的初始策略或生成模型来探索安全动作。

英文摘要:

In this paper, we study the problem of ensuring safety with a few shots of samples for partially unknown systems. We first characterize a fundamental limit when producing safe actions is not possible due to insufficient information or samples. Then, we develop a technique that can generate provably safe actions and recovery behaviors using a minimum number of samples. In the performance analysis, we also establish Nagumos theorem - like results with relaxed assumptions, which is potentially useful in other contexts. Finally, we discuss how the proposed method can be integrated into a policy gradient algorithm to assure safety and stability with a handful of samples without stabilizing initial policies or generative models to probe safe actions.

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。

↑