Active Learning of Dynamics for Data-Driven Control Using Koopman Operators
Active Learning of Dynamics for Data-Driven Control Using Koopman Operators
复制标题
DOI：
10.1109/tro.2019.2923880
复制
发表时间：
2019-10-01
影响因子：
7.8
通讯作者：
Murphey, Todd D.
中科院分区：
文献类型：
--
作者：
Abraham, Ian;Murphey, Todd D.
本文提出了一种主动学习策略的机器人系统，考虑到任务信息，使快速学习，并允许控制很容易地综合利用Koopman算子表示。我们首先激励使用表示线性Koopman算子系统的非线性系统，说明了改进的基于模型的控制性能与驱动货车der Pol系统。信息理论的方法，然后应用到Koopman运营商制定的动力系统，我们得到一个控制器的机器人动力学的主动学习。主动学习控制器的示出，以增加有关的Koopman运营商的信息的速率。此外，我们的主动学习控制器可以很容易地将建立在Koopman动态的政策，使快速主动学习和改进的控制的好处。使用四轴飞行器的结果说明了自由落体过程中的单执行主动学习和稳定能力。主动学习的结果扩展为自动化Koopman可观的，我们实现我们的方法对真实的机器人系统。
This paper presents an active learning strategy for robotic systems that takes into account task information, enables fast learning, and allows control to be readily synthesized by taking advantage of the Koopman operator representation. We first motivate the use of representing nonlinear systems as linear Koopman operator systems by illustrating the improved model-based control performance with an actuated Van der Pol system. Information-theoretic methods are then applied to the Koopman operator formulation of dynamical systems where we derive a controller for active learning of robot dynamics. The active learning controller is shown to increase the rate of information about the Koopman operator. In addition, our active learning controller can readily incorporate policies built on the Koopman dynamics, enabling the benefits of fast active learning and improved control. Results using a quadcopter illustrate single-execution active learning and stabilization capabilities during free fall. The results for active learning are extended for automating Koopman observables and we implement our method on real robotic systems.