Training Neural Network Controllers Using Control Barrier Functions in the Presence of Disturbances
Training Neural Network Controllers Using Control Barrier Functions in the Presence of Disturbances
复制标题
DOI：
10.1109/itsc45102.2020.9294485
复制
发表时间：
2020-01
期刊：
影响因子：
--
通讯作者：
Shakiba Yaghoubi;Georgios Fainekos;S. Sankaranarayanan
中科院分区：
文献类型：
--
作者：
Shakiba Yaghoubi;Georgios Fainekos;S. Sankaranarayanan
近年来，控制障碍函数（CBF）已被用于非线性系统可证明安全的反馈控制律的设计。这些反馈控制方法通常通过求解在线二次规划（QP）来计算下一个控制输入。对于资源受限的系统，实时求解QP可能是计算上昂贵的过程。在存在干扰的情况下，找到基于CBF的安全控制输入可能会花费更多的时间，因为找到干扰的最坏情况通常需要求解非线性规划。在这项工作中，我们建议使用模仿学习学习神经网络的反馈控制器，这将满足CBF约束。在此过程中，我们还发展了一类新的高阶CBF系统的外部干扰。我们证明了一个受到外部干扰的单周期模型的框架，例如，风或水流。
Control Barrier Functions (CBF) have been recently utilized in the design of provably safe feedback control laws for nonlinear systems. These feedback control methods typically compute the next control input by solving an online Quadratic Program (QP). Solving QPs in real-time can be a computationally expensive process for resource-constrained systems. In the presence of disturbances, finding CBF-based safe control inputs can get even more time consuming as finding the worst-case of the disturbance requires solving a nonlinear program in general. In this work, we propose to use imitation learning to learn Neural Network based feedback controllers which will satisfy the CBF constraints. In the process, we also develop a new class of High Order CBF for systems under external disturbances. We demonstrate the framework on a unicycle model subject to external disturbances, e.g., wind or currents.