Hierarchical Proportion Models for Motion Generation via Integration of Motion Primitives

2026-02-03

AI Summary

This work addresses the limitations of traditional imitation learning in handling complex, long-horizon tasks—namely its reliance on large amounts of high-quality demonstration data and poor adaptability. The authors propose a hierarchical imitation learning framework that enables efficient action generation through proportional blending of motion primitives. In this architecture, a high-level policy performs long-term planning, while a low-level policy learns multiple motion primitives and combines them in adjustable proportions to synthesize novel actions. Three variants of the proportion model—learnable, sampling-based, and replay-based—are introduced to balance flexibility, computational cost, and adaptability. Experiments on a real robotic pick-and-place task demonstrate that the approach successfully generates complex actions not present in the primitive set, with the sampling-based and replay-based models significantly outperforming standard hierarchical baselines in both stability and generalization.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Manipulation
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Imitation learning (IL) enables robots to acquire human-like motion skills from demonstrations, but it still requires extensive high-quality data and retraining to handle complex or long-horizon tasks. To improve data efficiency and adaptability, this study proposes a hierarchical IL framework that integrates motion primitives with proportion-based motion synthesis. The proposed method employs a two-layer architecture, where the upper layer performs long-term planning, while a set of lower-layer models learn individual motion primitives, which are combined according to specific proportions. Three model variants are introduced to explore different trade-offs between learning flexibility, computational cost, and adaptability: a learning-based proportion model, a sampling-based proportion model, and a playback-based proportion model, which differ in how the proportions are determined and whether the upper layer is trainable. Through real-robot pick-and-place experiments, the proposed models successfully generated complex motions not included in the primitive set. The sampling-based and playback-based proportion models achieved more stable and adaptable motion generation than the standard hierarchical model, demonstrating the effectiveness of proportion-based motion integration for practical robot learning.

Problem

• imitation learning
• motion primitives
• hierarchical models
• motion generation
• data efficiency

Innovation

• hierarchical imitation learning
• motion primitives
• proportion-based motion synthesis
• robot motion generation
• data-efficient learning