Hierarchical Intention-Aware Expressive Motion Generation for Humanoid Robots

2025-06-02

AI Summary

In real-time human–robot interaction with humanoid robots, intent recognition remains inaccurate, expressive motion generation lacks social appropriateness, and computational efficiency is insufficient. Method: This paper proposes a hierarchical intent-driven motion synthesis framework integrating contextual learning and a latent-space diffusion model. It introduces a novel hierarchical intent refinement mechanism—incorporating structured prompting, confidence estimation, social context awareness, and safety-aware fallback—to enable dynamic intent correction and adaptive response. A lightweight latent-space diffusion model, pre-trained on large-scale motion data and embedded with physical constraints and social norm priors, generates expressive motions efficiently. Contribution/Results: Evaluated on a physical robot platform, the method achieves real-time synthesis of highly diverse, physically plausible, and socially aligned gestures. It significantly improves interaction naturalness and robustness under dynamic, unstructured human input.

Technology Category

• Intelligent Robots: Human-Robot Interaction
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search

Abstract

Effective human-robot interaction requires robots to identify human intentions and generate expressive, socially appropriate motions in real-time. Existing approaches often rely on fixed motion libraries or computationally expensive generative models. We propose a hierarchical framework that combines intention-aware reasoning via in-context learning (ICL) with real-time motion generation using diffusion models. Our system introduces structured prompting with confidence scoring, fallback behaviors, and social context awareness to enable intention refinement and adaptive response. Leveraging large-scale motion datasets and efficient latent-space denoising, the framework generates diverse, physically plausible gestures suitable for dynamic humanoid interactions. Experimental validation on a physical platform demonstrates the robustness and social alignment of our method in realistic scenarios.

Problem

• Real-time expressive motion generation for humanoid robots
• Intention-aware reasoning with adaptive social responses
• Efficient diverse motion synthesis using diffusion models

Innovation

• Hierarchical framework combines ICL and diffusion models
• Structured prompting with confidence scoring and fallback
• Efficient latent-space denoising for diverse gestures