Approximate non-linear model predictive control with safety-augmented neural networks

TL;DR

This paper introduces a neural network-based approximation method for nonlinear model predictive control that guarantees safety and constraint satisfaction, enabling faster computation suitable for resource-limited systems.

Contribution

It proposes a safety-augmented neural network approach that ensures deterministic safety and constraint satisfaction in approximate MPC, with verification and fallback mechanisms.

Findings

Achieves significant computational speedups over traditional online optimization.

Guarantees deterministic safety and constraint satisfaction despite approximation errors.

Demonstrates effectiveness on nonlinear MPC benchmarks with resource-constrained systems.

Abstract

Model predictive control (MPC) achieves stability and constraint satisfaction for general nonlinear systems, but requires computationally expensive online optimization. This paper studies approximations of such MPC controllers via neural networks (NNs) to achieve fast online evaluation. We propose safety augmentation that yields deterministic guarantees for convergence and constraint satisfaction despite approximation inaccuracies. We approximate the entire input sequence of the MPC with NNs, which allows us to verify online if it is a feasible solution to the MPC problem. We replace the NN solution by a safe candidate based on standard MPC techniques whenever it is infeasible or has worse cost. Our method requires a single evaluation of the NN and forward integration of the input sequence online, which is fast to compute on resource-constrained systems. The proposed control framework is illustrated using two numerical non-linear MPC benchmarks of different complexity, demonstrating computational speedups that are orders of magnitude higher than online optimization. In the examples, we achieve deterministic safety through the safety-augmented NNs, where a naive NN implementation fails.