TransformerMPC: Accelerating Model Predictive Control via Transformers

TL;DR

TransformerMPC leverages transformer attention mechanisms to significantly accelerate MPC computations in robotic control, reducing runtime by up to 35 times while maintaining optimal constraint satisfaction.

Contribution

It introduces a novel transformer-based approach for online constraint removal and warm start initialization in MPC, enhancing computational efficiency without sacrificing accuracy.

Findings

Achieves up to 35x speedup in MPC computation

Maintains constraint satisfaction after constraint removal

Seamlessly integrates with existing MPC solvers

Abstract

In this paper, we address the problem of reducing the computational burden of Model Predictive Control (MPC) for real-time robotic applications. We propose TransformerMPC, a method that enhances the computational efficiency of MPC algorithms by leveraging the attention mechanism in transformers for both online constraint removal and better warm start initialization. Specifically, TransformerMPC accelerates the computation of optimal control inputs by selecting only the active constraints to be included in the MPC problem, while simultaneously providing a warm start to the optimization process. This approach ensures that the original constraints are satisfied at optimality. TransformerMPC is designed to be seamlessly integrated with any MPC solver, irrespective of its implementation. To guarantee constraint satisfaction after removing inactive constraints, we perform an offline verification to ensure that the optimal control inputs generated by the MPC solver meet all constraints. The effectiveness of TransformerMPC is demonstrated through extensive numerical simulations on complex robotic systems, achieving up to 35x improvement in runtime without any loss in performance.