Algorithmic design and implementation considerations of deep MPC

TL;DR

This paper discusses the design and implementation challenges of Deep Model Predictive Control, which combines neural networks with MPC to adaptively learn system uncertainties while ensuring safety.

Contribution

It introduces an approach for distributing control between neural networks and MPC, and analyzes how control authority distribution affects performance.

Findings

Proper control authority distribution is crucial for optimal performance.

Poor distribution can lead to subpar control outcomes.

Numerical experiments demonstrate the impact of control authority choices.

Abstract

Deep Model Predictive Control (Deep MPC) is an evolving field that integrates model predictive control and deep learning. This manuscript is focused on a particular approach, which employs deep neural network in the loop with MPC. This class of approaches distributes control authority between a neural network and an MPC controller, in such a way that the neural network learns the model uncertainties while the MPC handles constraints. The approach is appealing because training data collected while the system is in operation can be used to fine-tune the neural network, and MPC prevents unsafe behavior during those learning transients. This manuscript explains implementation challenges of Deep MPC, algorithmic way to distribute control authority and argues that a poor choice in distributing control authority may lead to poor performance. A reason of poor performance is explained through a numerical experiment on a four-wheeled skid-steer dynamics.