Constrained Deep Learning Based Nonlinear Model Predictive Control

TL;DR

This paper introduces a constrained deep neural network approach for learning nonlinear model predictive control policies, ensuring constraint satisfaction, stability, and probabilistic guarantees, demonstrated on a Furuta pendulum.

Contribution

It proposes a novel constrained DNN design for MPC that enforces constraints during training and provides stability and feasibility guarantees for nonlinear systems.

Findings

Superior control performance compared to traditional MPC.

Effective enforcement of constraints through constrained training.

Empirical guarantees of feasibility and optimality.

Abstract

Learning-based model predictive control (MPC) is an approach designed to reduce the computational cost of MPC. In this paper, a constrained deep neural network (DNN) design is proposed to learn MPC policy for nonlinear systems. Using constrained training of neural networks, MPC constraints are enforced effectively. Furthermore, recursive feasibility and robust stability conditions are derived for the learning-based MPC approach. Additionally, probabilistic feasibility and optimality empirical guarantees are provided for the learned control policy. The proposed algorithm is implemented on the Furuta pendulum and control performance is demonstrated and compared with the exact MPC and the normally trained learning-based MPC. The results show superior control performance and constraint satisfaction of the proposed approach.