Safe Optimal Control Using Stochastic Barrier Functions and Deep Forward-Backward SDEs
Safe Optimal Control Using Stochastic Barrier Functions and Deep Forward-Backward SDEs
复制标题
DOI：
--
复制
发表时间：
2020-09
期刊：
影响因子：
--
通讯作者：
M. Pereira;Ziyi Wang;Ioannis Exarchos;Evangelos A. Theodorou
中科院分区：
文献类型：
--
作者：
M. Pereira;Ziyi Wang;Ioannis Exarchos;Evangelos A. Theodorou
本文介绍了随机最优控制和随机动态优化的一种新公式，它保证了状态和控制约束下的安全性。该方法综合了正反向随机微分方程、随机屏障函数、可微凸优化和深度学习等概念。利用上述概念，设计了一种用于安全轨迹优化的神经网络体系结构，其中学习可以以端到端的方式执行。在三个系统上进行了仿真，验证了该方法的有效性。
This paper introduces a new formulation for stochastic optimal control and stochastic dynamic optimization that ensures safety with respect to state and control constraints. The proposed methodology brings together concepts such as Forward-Backward Stochastic Differential Equations, Stochastic Barrier Functions, Differentiable Convex Optimization and Deep Learning. Using the aforementioned concepts, a Neural Network architecture is designed for safe trajectory optimization in which learning can be performed in an end-to-end fashion. Simulations are performed on three systems to show the efficacy of the proposed methodology.