Hierarchical Hybrid Learning for Long-Horizon Contact-Rich Robotic Assembly

2024-09-24

arXiv.org

AI Summary

End-to-end imitation learning for long-horizon, multi-contact robotic precision assembly suffers from heavy reliance on extensive expert demonstrations and insufficient accuracy, while reinforcement learning exhibits low sample efficiency and struggles with long-sequence decision-making. Method: This paper proposes ARCH, a hierarchical hybrid learning framework. At the high level, a few-shot demonstration-driven imitation learning policy enables generalized skill scheduling; at the low level, deep reinforcement learning is unified with model predictive control to construct a continuous, parameterized skill primitive library, balancing precision and robustness. Contribution/Results: Evaluated on a real robotic platform, ARCH achieves zero-shot generalization from a single trained task to unseen assembly tasks. It significantly outperforms baseline methods in success rate and improves data efficiency by over 3×, demonstrating strong adaptability and scalability for complex, contact-rich manipulation.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Manipulation
• 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
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

Generalizable long-horizon robotic assembly requires reasoning at multiple levels of abstraction. End-to-end imitation learning (IL) has been proven a promising approach, but it requires a large amount of demonstration data for training and often fails to meet the high-precision requirement of assembly tasks. Reinforcement Learning (RL) approaches have succeeded in high-precision assembly tasks, but suffer from sample inefficiency and hence, are less competent at long-horizon tasks. To address these challenges, we propose a hierarchical modular approach, named ARCH (Adaptive Robotic Composition Hierarchy), which enables long-horizon high-precision assembly in contact-rich settings. ARCH employs a hierarchical planning framework, including a low-level primitive library of continuously parameterized skills and a high-level policy. The low-level primitive library includes essential skills for assembly tasks, such as grasping and inserting. These primitives consist of both RL and model-based controllers. The high-level policy, learned via imitation learning from a handful of demonstrations, selects the appropriate primitive skills and instantiates them with continuous input parameters. We extensively evaluate our approach on a real robot manipulation platform. We show that while trained on a single task, ARCH generalizes well to unseen tasks and outperforms baseline methods in terms of success rate and data efficiency. Videos can be found at https://long-horizon-assembly.github.io.

Problem

• Addressing high-precision robotic assembly generalization
• Overcoming data inefficiency in imitation and reinforcement learning
• Enabling long-horizon contact-rich task performance

Innovation

• Hierarchical planning framework with RL and model-based primitives
• High-level IL policy selects and parameterizes skills
• Combines imitation and reinforcement learning for assembly