Neural ODE-based Imitation Learning (NODE-IL): Data-efficient imitation learning for long-horizon multi-skill robot manipulation

Shiyao Zhao, Yucheng Xu, Mohammadreza Kasaei, Mohsen Khadem, Zhibin Li

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In robotics, acquiring new skills through Imitation Learning (IL) is crucial for handling diverse complex tasks. However, model-free IL faces challenges of data inefficiency and prolonged training time, whereas model-based methods struggle to obtain accurate nonlinear models. To address these challenges, we developed Neural ODE-based Imitation Learning (NODE-IL), a novel model-based imitation learning framework that employs Neural Ordinary Differential Equations (Neural ODEs) for learning task dynamics and control policies. NODE-IL comprises (1) Dynamic-NODE for learning the continuous differentiable task's transition dynamics model, and (2) Control-NODE for learning a long-horizon control policy in an MPC fashion, which are trained holistically. Extensively evaluated on challenging manipulation tasks, NODE-IL demonstrates significant advantages in data efficiency, requiring less than 70 samples to achieve robust performance. It outperforms Behavioral Cloning from Observation (BCO) and Gaussian Process Imitation Learning (GP-IL) methods, achieving 70% higher average success rate, and reducing translation errors for high-precision tasks, which demonstrates its robustness and accuracy, as an effective and efficient imitation learning approach for learning complex manipulation tasks.
Original languageEnglish
Title of host publicationProceedings of the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems
Publication statusAccepted/In press - 30 Jun 2024
Event2024 IEEE/RSJ International Conference on Intelligent Robots and Systems - Abu Dhabi, United Arab Emirates
Duration: 14 Oct 202418 Oct 2024
https://iros2024-abudhabi.org/

Conference

Conference2024 IEEE/RSJ International Conference on Intelligent Robots and Systems
Abbreviated titleIROS 2024
Country/TerritoryUnited Arab Emirates
CityAbu Dhabi
Period14/10/2418/10/24
Internet address

Fingerprint

Dive into the research topics of 'Neural ODE-based Imitation Learning (NODE-IL): Data-efficient imitation learning for long-horizon multi-skill robot manipulation'. Together they form a unique fingerprint.

Cite this