Approximate Robust NMPC using Reinforcement Learning

TL;DR

This paper introduces an RL-based approach to enhance robust nonlinear model predictive control, effectively managing uncertainties and disturbances with low computational complexity, demonstrated on a mobile robot simulation.

Contribution

It proposes a novel RL-RNMPC framework that improves robustness and performance of NMPC using ellipsoidal uncertainty modeling and reinforcement learning.

Findings

Successfully applied to a Wheeled Mobile Robot simulation.

Achieved improved trajectory tracking and obstacle avoidance.

Reduced computational complexity of robust NMPC.

Abstract

We present a Reinforcement Learning-based Robust Nonlinear Model Predictive Control (RL-RNMPC) framework for controlling nonlinear systems in the presence of disturbances and uncertainties. An approximate Robust Nonlinear Model Predictive Control (RNMPC) of low computational complexity is used in which the state trajectory uncertainty is modelled via ellipsoids. Reinforcement Learning is then used in order to handle the ellipsoidal approximation and improve the closed-loop performance of the scheme by adjusting the MPC parameters generating the ellipsoids. The approach is tested on a simulated Wheeled Mobile Robot (WMR) tracking a desired trajectory while avoiding static obstacles.