Deep Neural Network Approximation of Nonlinear Model Predictive Control. Issue 2 (2020)

• Record Type: Journal Article
• Authors: Cao, Yankai; Gopaluni, R. Bhushan
• Is Part Of: IFAC-PapersOnLine. Volume 53:Issue 2(2020)
• Page Start: 11319
• Page End: 11324
• Publication Date: 2020
• Subjects: Model Predictive Control -- Stochastic Optimization -- Deep Neural Networks -- Nonlinear Systems; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2020.12.538
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: This paper focuses on developing effective computational methods to enable the real-time application of model predictive control (MPC) for nonlinear systems. To achieve this goal, we follow the idea of approximating the MPC control law with a Deep Neural Network (DNN). To train the deep neural network offline, we propose a new "optimize and train" method that combines the steps of data generation and neural network training into a single high-dimensional stochastic optimization problem. This approach directly optimizes the closed loop performance of the DNN controller over a finite horizon for a number of initial states. The large-scale optimization problem can be solved efficiently using parallel computing techniques. The benefits of this approach over the conventional "optimize then train" protocol is illustrated through numerical results.