Online Intention Prediction via Control-Informed Learning

2026-04-10

AI Summary

This work proposes a real-time state prediction method for autonomous systems whose target dynamics involve unknown parameters and time-varying intentions. The approach models intention as learnable parameters within the objective function and estimates them through an inverse optimal control framework augmented with a control-aware online learning mechanism. A sliding time window is incorporated to forget outdated information, enabling adaptive tracking of evolving intentions, while an efficient gradient computation strategy facilitates real-time parameter updates. Extensive simulations under varying noise levels and hardware experiments on a quadrotor platform demonstrate that the method achieves high prediction accuracy and strong robustness, confirming its effectiveness and practicality in dynamic and complex environments.

Technology Category

• Intelligent Robots: State Estimation
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy

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 A shifting horizon strategy discounts outdated information, while online control-informed learning enables efficient gradient computation and online parameter updates.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.

Problem

• intention prediction
• online learning
• inverse optimal control
• autonomous systems
• time-varying intention

Innovation

• online intention prediction
• inverse optimal control
• control-informed learning
• shifting horizon
• real-time adaptation