Temporal Logic Imitation: Learning Plan-Satisficing Motion Policies from Demonstrations
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913600225935360 |
|---|---|
| author | Wang, Yanwei Figueroa, Nadia Li, Shen Shah, Ankit Shah, Julie |
| author_facet | Wang, Yanwei Figueroa, Nadia Li, Shen Shah, Ankit Shah, Julie |
| contents | Learning from demonstration (LfD) has succeeded in tasks featuring a long time horizon. However, when the problem complexity also includes human-in-the-loop perturbations, state-of-the-art approaches do not guarantee the successful reproduction of a task. In this work, we identify the roots of this challenge as the failure of a learned continuous policy to satisfy the discrete plan implicit in the demonstration. By utilizing modes (rather than subgoals) as the discrete abstraction and motion policies with both mode invariance and goal reachability properties, we prove our learned continuous policy can simulate any discrete plan specified by a linear temporal logic (LTL) formula. Consequently, an imitator is robust to both task- and motion-level perturbations and guaranteed to achieve task success. Project page: https://yanweiw.github.io/tli/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2206_04632 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Temporal Logic Imitation: Learning Plan-Satisficing Motion Policies from Demonstrations Wang, Yanwei Figueroa, Nadia Li, Shen Shah, Ankit Shah, Julie Robotics Artificial Intelligence Formal Languages and Automata Theory Machine Learning Systems and Control Learning from demonstration (LfD) has succeeded in tasks featuring a long time horizon. However, when the problem complexity also includes human-in-the-loop perturbations, state-of-the-art approaches do not guarantee the successful reproduction of a task. In this work, we identify the roots of this challenge as the failure of a learned continuous policy to satisfy the discrete plan implicit in the demonstration. By utilizing modes (rather than subgoals) as the discrete abstraction and motion policies with both mode invariance and goal reachability properties, we prove our learned continuous policy can simulate any discrete plan specified by a linear temporal logic (LTL) formula. Consequently, an imitator is robust to both task- and motion-level perturbations and guaranteed to achieve task success. Project page: https://yanweiw.github.io/tli/ |
| title | Temporal Logic Imitation: Learning Plan-Satisficing Motion Policies from Demonstrations |
| topic | Robotics Artificial Intelligence Formal Languages and Automata Theory Machine Learning Systems and Control |
| url | https://arxiv.org/abs/2206.04632 |