论文标题
障碍认证的安全学习控制：当方面计划达到强化学习时
Barrier Certified Safety Learning Control: When Sum-of-Square Programming Meets Reinforcement Learning
论文作者
论文摘要
安全保证在许多工程实施中至关重要。强化学习提供了一种有用的方法来增强安全性。但是，增强学习算法不能完全保证对现实操作的安全性。为了解决这一问题，这项工作采用了对强化学习的控制障碍功能，并提出了一种补偿算法以完全维持安全性。具体而言，已经利用了一个方案的总和来搜索最佳控制器，并同时调整学习超级方案。因此，保证控制措施始终在安全区域内。提出的方法的有效性通过倒置模型证明。与基于二次编程的增强学习方法相比，我们的基于方案的总和计划的增强性学习表明了它的优势。
Safety guarantee is essential in many engineering implementations. Reinforcement learning provides a useful way to strengthen safety. However, reinforcement learning algorithms cannot completely guarantee safety over realistic operations. To address this issue, this work adopts control barrier functions over reinforcement learning, and proposes a compensated algorithm to completely maintain safety. Specifically, a sum-of-squares programming has been exploited to search for the optimal controller, and tune the learning hyperparameters simultaneously. Thus, the control actions are pledged to be always within the safe region. The effectiveness of proposed method is demonstrated via an inverted pendulum model. Compared to quadratic programming based reinforcement learning methods, our sum-of-squares programming based reinforcement learning has shown its superiority.
