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 2008.05952cs.LGcs.SYeess.SYstat.ML
收藏

从数据中学习稳定性证书

Learning Stability Certificates from Data

John A. Paulson School of Engineering and Applied Sciences, Harvard University(哈佛大学约翰·A·保尔森工程与应用科学学院);Google Brain Robotics(谷歌大脑机器人团队);Department of Electrical and Systems Engineering, University of Pennsylvania(宾夕法尼亚大学电气与系统工程系);Nonlinear Systems Laboratory, Massachusetts Institute of Technology(麻省理工学院非线性系统实验室)

Nicholas M. Boffi, Stephen Tu, Nikolai Matni, Jean-Jacques E. Slotine, Vikas Sindhwani

摘要:

非线性控制理论中用于建立动态系统的稳定性或安全性的许多现有工具可以被提炼为保证期望属性的证书函数的构造。然而,用于合成证书函数的算法通常需要底层动力学的封闭形式的分析表达式,这排除了它们在许多现代机器人平台上的使用。为了规避这个问题,我们开发的算法学习证书功能只从轨迹数据。我们建立的泛化误差的界限-证书将不会证明一个新的,看不见的轨迹的概率-当从轨迹学习,我们将这样的泛化误差界限转换为全局稳定性保证。我们的经验表明,证书复杂的动力学可以有效地学习,学习证书可用于下游任务,如自适应控制。

英文摘要:

Many existing tools in nonlinear control theory for establishing stability or safety of a dynamical system can be distilled to the construction of a certificate function that guarantees a desired property. However, algorithms for synthesizing certificate functions typically require a closed-form analytical expression of the underlying dynamics, which rules out their use on many modern robotic platforms. To circumvent this issue, we develop algorithms for learning certificate functions only from trajectory data. We establish bounds on the generalization error - the probability that a certificate will not certify a new, unseen trajectory - when learning from trajectories, and we convert such generalization error bounds into global stability guarantees. We demonstrate empirically that certificates for complex dynamics can be efficiently learned, and that the learned certificates can be used for downstream tasks such as adaptive control.

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。

↑