Safe and Efficient Model Predictive Control Using Neural Networks: An Interior Point Approach
Safe and Efficient Model Predictive Control Using Neural Networks: An Interior Point Approach
复制标题
DOI：
10.1109/cdc51059.2022.9993046
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Daniel Tabas;Baosen Zhang
中科院分区：
文献类型：
--
作者：
Daniel Tabas;Baosen Zhang
模型预测控制（MPC）为控制具有约束的系统提供了一种有用的手段，但是遭受真实的时间中重复求解优化问题的计算负担。MPC的离线（显式）解决方案试图使用多参数编程或机器学习来减轻真实的计算挑战。多参数方法通常应用于线性或二次MPC问题，而基于学习的方法可以更灵活，内存密集度更低。现有的基于学习的方法提供了显着的加速，但挑战成为确保约束满足，同时保持良好的性能。在本文中，我们提供了一个神经网络参数化的MPC政策，明确编码的问题的约束。通过在无监督学习范式中探索MPC可行集的内部，神经网络比基于投影的方法更快地找到更好的策略，并且具有更短的求解时间。我们使用所提出的政策来解决一个强大的MPC问题，并展示了一个标准的测试系统上的性能和计算增益。
Model predictive control (MPC) provides a useful means for controlling systems with constraints, but suffers from the computational burden of repeatedly solving an optimization problem in real time. Offline (explicit) solutions for MPC attempt to alleviate real time computational challenges using either multiparametric programming or machine learning. The multiparametric approaches are typically applied to linear or quadratic MPC problems, while learning-based approaches can be more flexible and are less memory-intensive. Existing learning-based approaches offer significant speedups, but the challenge becomes ensuring constraint satisfaction while maintaining good performance. In this paper, we provide a neural network parameterization of MPC policies that explicitly encodes the constraints of the problem. By exploring the interior of the MPC feasible set in an unsupervised learning paradigm, the neural network finds better policies faster than projectionbased methods and exhibits substantially shorter solve times. We use the proposed policy to solve a robust MPC problem, and demonstrate the performance and computational gains on a standard test system.