Learning Skill-based Industrial Robot Tasks with User Priors
(2022) IEEE 18th International Conference on Automation Science and Engineering (IEEE CASE2022) p.1485-1492- Abstract
- Robot skills systems are meant to reduce robot setup time for new manufacturing tasks. Yet, for dexterous, contact-rich tasks, it is often difficult to find the right skill parameters. One strategy is to learn these parameters by allowing the robot system to learn directly on the task. For a learning problem, a robot operator can typically specify the type and range of values of the parameters. Nevertheless, given their prior experience, robot operators should be able to help the learning process further by providing educated guesses about where in the parameter space potential optimal solutions could be found. Interestingly, such prior knowledge is not exploited in current robot learning frameworks. We introduce an approach that combines... (More)
- Robot skills systems are meant to reduce robot setup time for new manufacturing tasks. Yet, for dexterous, contact-rich tasks, it is often difficult to find the right skill parameters. One strategy is to learn these parameters by allowing the robot system to learn directly on the task. For a learning problem, a robot operator can typically specify the type and range of values of the parameters. Nevertheless, given their prior experience, robot operators should be able to help the learning process further by providing educated guesses about where in the parameter space potential optimal solutions could be found. Interestingly, such prior knowledge is not exploited in current robot learning frameworks. We introduce an approach that combines user priors and Bayesian optimization to allow fast optimization of robot industrial tasks at robot deployment time. We evaluate our method on three tasks that are learned in simulation as well as on two tasks that are learned directly on a real robot system. Additionally, we transfer knowledge from the corresponding simulation tasks by automatically constructing priors from well-performing configurations for learning on the real system. To handle potentially contradicting task objectives, the tasks are modeled as multi-objective problems. Our results show that operator priors, both user-specified and transferred, vastly accelerate the discovery of rich Pareto fronts, and typically produce final performance far superior to proposed baselines. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/f86902d3-1c05-4e7d-a1f1-0933b77223be
- author
- Mayr, Matthias
LU
; Hvarfner, Carl LU ; Chatzilygeroudis, Konstantinos ; Nardi, Luigi LU and Krueger, Volker LU
- organization
- publishing date
- 2022
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- IEEE International Conference on Automation Science and Engineering (CASE)
- pages
- 1485 - 1492
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- IEEE 18th International Conference on Automation Science and Engineering (IEEE CASE2022)
- conference location
- Mexico City, Mexico
- conference dates
- 2022-08-20 - 2022-08-24
- external identifiers
-
- scopus:85141671408
- ISBN
- 978-1-6654-9043-6
- 978-1-6654-9042-9
- DOI
- 10.1109/CASE49997.2022.9926713
- project
- RobotLab LTH
- Robotics and Semantic Systems
- WASP Package: Bayesian optimization methods and their applications to real-world problems
- WASP Professor Package: Cognitive Robots for Manufacturing
- Efficient Learning of Robot Skills
- language
- English
- LU publication?
- yes
- id
- f86902d3-1c05-4e7d-a1f1-0933b77223be
- date added to LUP
- 2022-09-19 13:39:27
- date last changed
- 2024-09-05 23:56:52
@inproceedings{f86902d3-1c05-4e7d-a1f1-0933b77223be, abstract = {{Robot skills systems are meant to reduce robot setup time for new manufacturing tasks. Yet, for dexterous, contact-rich tasks, it is often difficult to find the right skill parameters. One strategy is to learn these parameters by allowing the robot system to learn directly on the task. For a learning problem, a robot operator can typically specify the type and range of values of the parameters. Nevertheless, given their prior experience, robot operators should be able to help the learning process further by providing educated guesses about where in the parameter space potential optimal solutions could be found. Interestingly, such prior knowledge is not exploited in current robot learning frameworks. We introduce an approach that combines user priors and Bayesian optimization to allow fast optimization of robot industrial tasks at robot deployment time. We evaluate our method on three tasks that are learned in simulation as well as on two tasks that are learned directly on a real robot system. Additionally, we transfer knowledge from the corresponding simulation tasks by automatically constructing priors from well-performing configurations for learning on the real system. To handle potentially contradicting task objectives, the tasks are modeled as multi-objective problems. Our results show that operator priors, both user-specified and transferred, vastly accelerate the discovery of rich Pareto fronts, and typically produce final performance far superior to proposed baselines.}}, author = {{Mayr, Matthias and Hvarfner, Carl and Chatzilygeroudis, Konstantinos and Nardi, Luigi and Krueger, Volker}}, booktitle = {{IEEE International Conference on Automation Science and Engineering (CASE)}}, isbn = {{978-1-6654-9043-6}}, language = {{eng}}, pages = {{1485--1492}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, title = {{Learning Skill-based Industrial Robot Tasks with User Priors}}, url = {{http://dx.doi.org/10.1109/CASE49997.2022.9926713}}, doi = {{10.1109/CASE49997.2022.9926713}}, year = {{2022}}, }