Aspproximating Model Predictive Controller With Biased ReLU Neural Network

Aspproximating Model Predictive Controller With Biased ReLU Neural Network
复制标题

利用偏置 ReLU 神经网络逼近模型预测控制器

DOI:
10.1109/cac51589.2020.9326497
复制
发表时间:
2020-11
期刊:
2020 Chinese Automation Congress (CAC)
影响因子:
--
通讯作者:
Jun Xu
Jun Xu
中科院分区:
其他
文献类型:
--
作者:
Kai Wang;Xinglong Liang;Jun Xu

文献摘要

参考文献

相似文献

利用神经网络对所设计的MPC控制器进行逼近,在保证系统稳定性和满足约束条件的前提下,减少计算量,实现对系统的控制,是一个热门的研究领域。在MPC控制器上采用了一种新的约束收紧策略,以保证在给定的界内对输入扰动具有鲁棒性,从而达到约束满足和递归可行的目的。训练一个神经网络通常是耗时的,特别是当激活函数是非线性的,这导致实现不切实际。为了减少训练时间,我们使用有偏ReLU神经网络来近似控制器。比较了有偏ReLU神经网络和全连接神经网络逼近MPC控制器的性能。有偏ReLU神经网络的训练时间要短得多,同时达到类似的准确度。
It is a popular research area to use neural network to approximate the designed MPC controller to control the system with reduced computational burden while maintaining stability and constraint satisfaction. A new strategy called constraints tightening is used on the MPC controller to ensure robustness to input disturbance within given bound in order that the constraints satisfaction and recursive feasibility can be achieved. Training a neural network is usually time-consuming especially when the activation function is nonlinear, which leads the implementation unpratical. To reduce the time for training we use a biased ReLU neural network to approximate the controller. The performance of the biased ReLU neural network and fully connected neural network in approximating the MPC controller is compared. The training time for the biased ReLU neural network is much shorter while achieving the similar level of accuracy.
DOI: 10.1016/s0005-1098(98)00073-9
发表时间: 1998-10-01
期刊: AUTOMATICA
影响因子: 6.4
作者:
Chen, H;Allgower, F
通讯作者: Allgower, F
DOI: 10.1109/tac.2018.2800789
发表时间: 2019-01
影响因子: 6.8
作者:
Johannes Köhler;M. Müller;F. Allgöwer
通讯作者: Johannes Köhler;M. Müller;F. Allgöwer
有偏差的 ReLU 神经网络
DOI: 10.1016/j.neucom.2020.09.050
发表时间: 2021-01-29
期刊: NEUROCOMPUTING
影响因子: 6
作者:
Liang, XingLong;Xu, Jun
通讯作者: Xu, Jun
DOI: 10.1016/j.automatica.2011.08.023
发表时间: 2011-10
期刊: Autom.
影响因子: --
作者:
M. Kvasnica;Johan Löfberg;M. Fikar
通讯作者: M. Kvasnica;Johan Löfberg;M. Fikar
DOI: 10.1002/rnc.1758
发表时间: 2011-07
影响因子: 3.9
作者:
D. Mayne;E. Kerrigan;E. V. Wyk;P. Falugi
通讯作者: D. Mayne;E. Kerrigan;E. V. Wyk;P. Falugi