Learning Nonlinear Systems In-Context: From Synthetic Data to Real-World Motor Control

TL;DR

This paper introduces a transformer-based in-context learning approach for nonlinear motor control, enabling models trained on synthetic data to adapt effectively to real-world systems with minimal examples.

Contribution

It is the first to apply in-context learning with transformers to signal processing systems, specifically for nonlinear motor control, bridging synthetic training and real-world application.

Findings

Model generalizes across various motor load configurations.

Outperforms traditional PI controllers and physics-based baselines.

Enables accurate one-shot predictions with minimal data.

Abstract

LLMs have shown strong in-context learning (ICL) abilities, but have not yet been extended to signal processing systems. Inspired by their design, we have proposed for the first time ICL using transformer models applicable to motor feedforward control, a critical task where classical PI and physics-based methods struggle with nonlinearities and complex load conditions. We propose a transformer based model architecture that separates signal representation from system behavior, enabling both few-shot finetuning and one-shot ICL. Pretrained on a large corpus of synthetic linear and nonlinear systems, the model learns to generalize to unseen system dynamics of real-world motors only with a handful of examples. In experiments, our approach generalizes across multiple motor load configurations, transforms untuned examples into accurate feedforward predictions, and outperforms PI controllers and physics-based feedforward baselines. These results demonstrate that ICL can bridge synthetic pretraining and real-world adaptability, opening new directions for data efficient control of physical systems.