End-to-End Reinforcement Learning of Koopman Models for Economic Nonlinear Model Predictive Control

Abstract

(Economic) nonlinear model predictive control ((e)NMPC) requires dynamic models that are sufficiently accurate and computationally tractable. Data-driven surrogate models for mechanistic models can reduce the computational burden of (e)NMPC; however, such models are typically trained by system identification for maximum prediction accuracy on simulation samples and perform suboptimally in (e)NMPC. We present a method for end-to-end reinforcement learning of Koopman surrogate models for optimal performance as part of (e)NMPC. We apply our method to two applications derived from an established nonlinear continuous stirred-tank reactor model. The controller performance is compared to that of (e)NMPCs utilizing models trained using system identification, and model-free neural network controllers trained using reinforcement learning. We show that the end-to-end trained models outperform those trained using system identification in (e)NMPC, and that, in contrast to the neural network controllers, the (e)NMPC controllers can react to changes in the control setting without retraining.

arXiv commentmanuscript (20 pages, 7 figures, 6 tables), supplementary materials (3 pages, 2 tables)

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• Addressing Function Approximation Error in Actor-Critic Methods
• High-Dimensional Continuous Control Using Generalized Advantage Estimation
• Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
• Deep learning for universal linear embeddings of nonlinear dynamics
• Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control
• Data-driven Economic NMPC using Reinforcement Learning
• Differentiable MPC for End-to-end Planning and Control
• Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPO
• OptNet: Differentiable Optimization as a Layer in Neural Networks
• A data-driven Koopman model predictive control framework for nonlinear flows
• Wild-Time: A Benchmark of in-the-Wild Distribution Shift over Time
• Accelerated Policy Learning with Parallel Differentiable Simulation
• Task-optimal data-driven surrogate models for eNMPC via differentiable simulation and optimization
• Data-driven End-to-end Learning of Pole Placement Control for Nonlinear Dynamics via Koopman Invariant Subspaces

Cited by in corpus (1)

• Computationally efficient Gauss-Newton reinforcement learning for model predictive control