English

Online Intention Prediction via Control-Informed Learning

Robotics 2026-04-13 v1 Machine Learning Systems and Control Systems and Control

Abstract

This paper presents an online intention prediction framework for estimating the goal state of autonomous systems in real time, even when intention is time-varying, and system dynamics or objectives include unknown parameters. The problem is formulated as an inverse optimal control / inverse reinforcement learning task, with the intention treated as a parameter in the objective. A shifting horizon strategy discounts outdated information, while online control-informed learning enables efficient gradient computation and online parameter updates. Simulations under varying noise levels and hardware experiments on a quadrotor drone demonstrate that the proposed approach achieves accurate, adaptive intention prediction in complex environments.

Keywords

Cite

@article{arxiv.2604.09303,
  title  = {Online Intention Prediction via Control-Informed Learning},
  author = {Tianyu Zhou and Zihao Liang and Zehui Lu and Shaoshuai Mou},
  journal= {arXiv preprint arXiv:2604.09303},
  year   = {2026}
}
R2 v1 2026-07-01T12:02:53.923Z