Projects per year
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 language | English |
---|---|
Title of host publication | Proceedings of the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems |
Publication status | Accepted/In press - 30 Jun 2024 |
Event | 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems - Abu Dhabi, United Arab Emirates Duration: 14 Oct 2024 → 18 Oct 2024 https://iros2024-abudhabi.org/ |
Conference
Conference | 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems |
---|---|
Abbreviated title | IROS 2024 |
Country/Territory | United Arab Emirates |
City | Abu Dhabi |
Period | 14/10/24 → 18/10/24 |
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.Projects
- 1 Finished
-
A Cyber-Physical System for Unified Diagnosis and Treatment of Lung Diseases
Khadem, M., Akram, A., Dhaliwal, K. & Ramamoorthy, R.
1/05/20 → 30/04/24
Project: Research