arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2304.08779eess.SYcs.SYmath.OC
收藏

模型预测控制的 maxout 神经网络逼近误差界

Error bounds for maxout neural network approximations of model predictive control

Dieter Teichrib, Moritz Schulze Darup

AI总结:

针对神经网络逼近 MPC 难以保留稳定性与性能保证的问题,提出用混合整数可行性问题精确计算 maxout NN 的最大误差与局部增益,并证明存在零误差逼近。

摘要:

当底层最优控制问题(OCP)的在线求解要求过高,且显式 MPC 控制律的精确计算难以处理时,采用神经网络(NN)逼近模型预测控制(MPC)是一种灵活通用的方法。此类逼近的缺点是通常无法保留原始 MPC 的稳定性与性能保证。然而,如果已知其相对于最优控制律的最大误差以及该误差的 Lipschitz 常数,这些保证便可恢复。本文展示了当控制律由 maxout NN 逼近时,如何精确计算这两个值。我们基于 ReLU NN 逼近的相关结果,推导混合整数(MI)线性约束,使得通过求解一个 MI 可行性问题即可计算 maxout NN 的输出和局部增益。此外,我们从理论和实验两方面证明,存在最大误差为零的 maxout NN。

英文摘要:

Neural network (NN) approximations of model predictive control (MPC) are a versatile approach if the online solution of the underlying optimal control problem (OCP) is too demanding and if an exact computation of the explicit MPC law is intractable. The drawback of such approximations is that they typically do not preserve stability and performance guarantees of the original MPC. However, such guarantees can be recovered if the maximum error with respect to the optimal control law and the Lipschitz constant of that error are known. We show in this work how to compute both values exactly when the control law is approximated by a maxout NN. We build upon related results for ReLU NN approximations and derive mixed-integer (MI) linear constraints that allow a computation of the output and the local gain of a maxout NN by solving an MI feasibility problem. Furthermore, we show theoretically and experimentally that maxout NN exist for which the maximum error is zero.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑