Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Skill-based Multi-objective Reinforcement Learning of Industrial Robot Tasks with Planning and Knowledge Integration

Mayr, Matthias LU orcid ; Ahmad, Faseeh LU ; Chatzilygeroudis, Konstantinos ; Nardi, Luigi LU and Krueger, Volker LU orcid (2022) 2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022 p.1995-2002
Abstract

In modern industrial settings with small batch sizes it should be easy to set up a robot system for a new task. Strategies exist, e.g. the use of skills, but when it comes to handling forces and torques, these systems often fall short. We introduce an approach that provides a combination of task-level planning with targeted learning of scenario-specific parameters for skill-based systems. We propose the following pipeline: the user provides a task goal in the planning language PDDL, then a plan (i.e., a sequence of skills) is generated and the learnable parameters of the skills are automatically identified, and, finally, an operator chooses reward functions and parameters for the learning process. Two aspects of our methodology are... (More)

In modern industrial settings with small batch sizes it should be easy to set up a robot system for a new task. Strategies exist, e.g. the use of skills, but when it comes to handling forces and torques, these systems often fall short. We introduce an approach that provides a combination of task-level planning with targeted learning of scenario-specific parameters for skill-based systems. We propose the following pipeline: the user provides a task goal in the planning language PDDL, then a plan (i.e., a sequence of skills) is generated and the learnable parameters of the skills are automatically identified, and, finally, an operator chooses reward functions and parameters for the learning process. Two aspects of our methodology are critical: (a) learning is tightly integrated with a knowledge framework to support symbolic planning and to provide priors for learning, (b) using multi-objective optimization. This can help to balance key performance indicators (KPIs) such as safety and task performance since they can often affect each other. We adopt a multi-objective Bayesian optimization approach and learn entirely in simulation. We demonstrate the efficacy and versatility of our approach by learning skill parameters for two different contact-rich tasks. We show their successful execution on a real 7-DOF KUKA-iiwa manipulator and outperform the manual parameterization by human robot operators.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Reinforcement Learning, Multi-objective, Knowledge Representation
host publication
2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022
pages
8 pages
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
conference name
2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022
conference location
Jinghong, China
conference dates
2022-12-05 - 2022-12-09
external identifiers
  • scopus:85147332734
ISBN
9781665481090
DOI
10.1109/ROBIO55434.2022.10011996
project
RobotLab LTH
Efficient Learning of Robot Skills
Robotics and Semantic Systems
WASP Professor Package: Cognitive Robots for Manufacturing
language
English
LU publication?
yes
id
57122c71-3201-4d86-85a9-00b3b51935f3
date added to LUP
2023-02-20 15:18:44
date last changed
2024-03-06 08:44:30
@inproceedings{57122c71-3201-4d86-85a9-00b3b51935f3,
  abstract     = {{<p>In modern industrial settings with small batch sizes it should be easy to set up a robot system for a new task. Strategies exist, e.g. the use of skills, but when it comes to handling forces and torques, these systems often fall short. We introduce an approach that provides a combination of task-level planning with targeted learning of scenario-specific parameters for skill-based systems. We propose the following pipeline: the user provides a task goal in the planning language PDDL, then a plan (i.e., a sequence of skills) is generated and the learnable parameters of the skills are automatically identified, and, finally, an operator chooses reward functions and parameters for the learning process. Two aspects of our methodology are critical: (a) learning is tightly integrated with a knowledge framework to support symbolic planning and to provide priors for learning, (b) using multi-objective optimization. This can help to balance key performance indicators (KPIs) such as safety and task performance since they can often affect each other. We adopt a multi-objective Bayesian optimization approach and learn entirely in simulation. We demonstrate the efficacy and versatility of our approach by learning skill parameters for two different contact-rich tasks. We show their successful execution on a real 7-DOF KUKA-iiwa manipulator and outperform the manual parameterization by human robot operators.</p>}},
  author       = {{Mayr, Matthias and Ahmad, Faseeh and Chatzilygeroudis, Konstantinos and Nardi, Luigi and Krueger, Volker}},
  booktitle    = {{2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022}},
  isbn         = {{9781665481090}},
  keywords     = {{Reinforcement Learning; Multi-objective; Knowledge Representation}},
  language     = {{eng}},
  pages        = {{1995--2002}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  title        = {{Skill-based Multi-objective Reinforcement Learning of Industrial Robot Tasks with Planning and Knowledge Integration}},
  url          = {{http://dx.doi.org/10.1109/ROBIO55434.2022.10011996}},
  doi          = {{10.1109/ROBIO55434.2022.10011996}},
  year         = {{2022}},
}