Behavior Foundation Model: Towards Next-Generation Whole-Body Control System of Humanoid Robots

2025-06-25

AI Summary

To address the challenges of whole-body control (WBC) for humanoid robots—including high dynamical complexity, underactuation, and limited adaptability to diverse tasks—as well as the poor generalization and task-specific retraining requirements of existing learning-based controllers, this paper introduces the Behavior Foundation Model (BFM) paradigm. BFM leverages large-scale real-world and simulated behavioral data for pretraining, integrating motion policy modeling, behavioral decomposition, and a modular architecture to learn transferable primitive skills and behavior priors. The resulting model enables zero-shot cross-task transfer and few-shot rapid adaptation, substantially reducing downstream control customization effort. Furthermore, the work establishes a systematic development roadmap and open-sources an evolving BFM paper repository and toolkit. Collectively, this provides a unified, scalable, embodied intelligence foundation for humanoid robot behavior.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Humanoid robots are drawing significant attention as versatile platforms for complex motor control, human-robot interaction, and general-purpose physical intelligence. However, achieving efficient whole-body control (WBC) in humanoids remains a fundamental challenge due to sophisticated dynamics, underactuation, and diverse task requirements. While learning-based controllers have shown promise for complex tasks, their reliance on labor-intensive and costly retraining for new scenarios limits real-world applicability. To address these limitations, behavior(al) foundation models (BFMs) have emerged as a new paradigm that leverages large-scale pretraining to learn reusable primitive skills and behavioral priors, enabling zero-shot or rapid adaptation to a wide range of downstream tasks. In this paper, we present a comprehensive overview of BFMs for humanoid WBC, tracing their development across diverse pre-training pipelines. Furthermore, we discuss real-world applications, current limitations, urgent challenges, and future opportunities, positioning BFMs as a key approach toward scalable and general-purpose humanoid intelligence. Finally, we provide a curated and long-term list of BFM papers and projects to facilitate more subsequent research, which is available at https://github.com/yuanmingqi/awesome-bfm-papers.

Problem

• Achieving efficient whole-body control in humanoid robots
• Reducing reliance on labor-intensive retraining for new scenarios
• Enabling zero-shot or rapid adaptation to diverse tasks

Innovation

• Leverages large-scale pretraining for reusable skills
• Enables zero-shot or rapid task adaptation
• Provides comprehensive BFM overview and resources