Energy-based Regularization for Learning Residual Dynamics in Neural MPC for Omnidirectional Aerial Robots

TL;DR

This paper introduces an energy-based regularization for neural residual dynamics in MPC, enhancing stability and accuracy in omnidirectional aerial robots by embedding physical energy considerations into learning.

Contribution

It proposes a novel energy regularization loss for neural residual dynamics, improving stability and accuracy in neural MPC for aerial robots.

Findings

Energy regularization reduces MAE by 23% in real-world tests.

Method improves flight stability compared to standard neural MPC.

Code is publicly available at the provided GitHub link.

Abstract

Data-driven Model Predictive Control (MPC) has lately been the core research subject in the field of control theory. The combination of an optimal control framework with deep learning paradigms opens up the possibility to accurately track control tasks without the need for complex analytical models. However, the system dynamics are often nuanced and the neural model lacks the potential to understand physical properties such as inertia and conservation of energy. In this work, we propose a novel energy-based regularization loss function which is applied to the training of a neural model that learns the residual dynamics of an omnidirectional aerial robot. Our energy-based regularization encourages the neural network to cause control corrections that stabilize the energy of the system. The residual dynamics are integrated into the MPC framework and improve the positional mean absolute error (MAE) over three real-world experiments by 23% compared to an analytical MPC. We also compare our method to a standard neural MPC implementation without regularization and primarily achieve a significantly increased flight stability implicitly due to the energy regularization and up to 15% lower MAE. Our code is available under: https://github.com/johanneskbl/jsk_aerial_robot/tree/develop/neural_MPC.