Learning Skill-based Industrial Robot Tasks with User Priors

Matthias Mayr, Carl Hvarfner, Konstantinos Chatzilygeroudis, Luigi Nardi, Volker Krueger

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingpeer-review

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.
Original languageEnglish
Title of host publicationIEEE International Conference on Automation Science and Engineering (CASE)
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
Pages1485-1492
ISBN (Electronic)978-1-6654-9042-9
ISBN (Print)978-1-6654-9043-6
DOIs
Publication statusPublished - 2022
EventIEEE 18th International Conference on Automation Science and Engineering (IEEE CASE2022) - Mexico City, Mexico
Duration: 2022 Aug 202022 Aug 24
Conference number: 18th
https://www.case2022.org/

Conference

ConferenceIEEE 18th International Conference on Automation Science and Engineering (IEEE CASE2022)
Country/TerritoryMexico
CityMexico City
Period2022/08/202022/08/24
Internet address

Subject classification (UKÄ)

  • Robotics

Fingerprint

Dive into the research topics of 'Learning Skill-based Industrial Robot Tasks with User Priors'. Together they form a unique fingerprint.

Cite this