Academic Journal
Learning a flexible neural energy function with a unique minimum for globally stable and accurate demonstration learning
| العنوان: | Learning a flexible neural energy function with a unique minimum for globally stable and accurate demonstration learning |
|---|---|
| المؤلفون: | Jin, Zhehao, Si, Weiyong, Liu, Andong, Zhang, Wen An, Yu, Li, Yang, Chenguang |
| بيانات النشر: | Institute of Electrical and Electronics Engineers |
| سنة النشر: | 2023 |
| المجموعة: | University of the West of England, Bristol: UWE Research Repository |
| مصطلحات موضوعية: | Electrical and Electronic Engineering, Computer Science Applications, Control and Systems Engineering |
| الوصف: | Learning a stable autonomous dynamic system (ADS) encoding human motion rules has been shown as an effective way for demonstration learning. However, the stability guarantee may sacrifice the demonstration learning accuracy. This article solves the issue by learning a stability certificate, represented by a neural energy function, on the demonstration set. We propose a polarlike space analysis approach to derive parameter constraints to guarantee the unique-minimum property of the neural energy function, which is essential for it to be a cogent stability certificate. Then, the neural energy function is learned to capture the demonstration preferences via constrained optimization algorithms. With the learned neural energy function, a globally asymptotically stable ADS with predefined position constraint is further formulated. We also quantitatively analyze the generalization ability of the learned ADS by utilizing the substantial flexibility of the neural energy function. The effectiveness of the proposed approach is validated on the LASA dataset and two representative robotic experiments. |
| نوع الوثيقة: | article in journal/newspaper |
| اللغة: | English |
| تدمد: | 1552-3098 |
| Relation: | https://uwe-repository.worktribe.com/file/11086298/1/Learning%20a%20flexible%20neural%20energy%20function%20with%20a%20unique%20minimum%20for%20globally%20stable%20and%20accurate%20demonstration%20learning; https://uwe-repository.worktribe.com/output/11086298 |
| DOI: | 10.1109/tro.2023.3303011 |
| الاتاحة: | https://uwe-repository.worktribe.com/output/11086298 https://doi.org/10.1109/tro.2023.3303011 |
| Rights: | openAccess ; http://www.rioxx.net/licenses/all-rights-reserved |
| رقم الانضمام: | edsbas.F6F2A6C7 |
| قاعدة البيانات: | BASE |
| تدمد: | 15523098 |
|---|---|
| DOI: | 10.1109/tro.2023.3303011 |