Adaptive Nonlinear Model Predictive Control for a Real-World Labyrinth Game

TL;DR

This paper introduces an adaptive nonlinear model predictive control framework for a real-world labyrinth game, combining high-level planning with low-level real-time control, and learning surface irregularities for improved robustness.

Contribution

It proposes a novel two-layer MPC approach with adaptive constraints and surface learning, enhancing control performance in complex, real-world labyrinth scenarios.

Findings

Outperforms classical control methods in robustness.

Effectively handles disturbances and model inaccuracies.

Learns surface irregularities to improve control accuracy.

Abstract

We present a nonlinear non-convex model predictive control approach to solving a real-world labyrinth game. We introduce adaptive nonlinear constraints, representing the non-convex obstacles within the labyrinth. Our method splits the computation-heavy optimization problem into two layers; first, a high-level model predictive controller which incorporates the full problem formulation and finds pseudo-global optimal trajectories at a low frequency. Secondly, a low-level model predictive controller that receives a reduced, computationally optimized version of the optimization problem to follow the given high-level path in real-time. Further, a map of the labyrinth surface irregularities is learned. Our controller is able to handle the major disturbances and model inaccuracies encountered on the labyrinth and outperforms other classical control methods.