Vision-Based Machine Learning in Robot Soccer

J.J. Olthuis; Noah van der Meer; S.T. Kempers; C.A. van Hoof; R.M. Beumer; W.J.P. Kuijpers; A.A. Kokkelmans; Wouter Houtman; J.J.F.J. van Eijck; J.J. Kon; A.T.A. Peijnenburg; M.J.G. van de Molengraft

doi:10.1007/978-3-030-98682-7_27

Vision-Based Machine Learning in Robot Soccer

J.J. Olthuis
, Noah van der Meer
, S.T. Kempers
, C.A. van Hoof
, R.M. Beumer
, W.J.P. Kuijpers
, A.A. Kokkelmans
, Wouter Houtman
, J.J.F.J. van Eijck
, J.J. Kon
, A.T.A. Peijnenburg
, M.J.G. van de Molengraft

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

4 Downloads (Pure)

Abstract

Robots need to perceive their environment in order to properly interact with it. In the RoboCup Soccer Middle Size League (MSL) this happens primarily through cameras mounted on the robots. Machine Learning can be used to extract relevant features from camera imagery. The real-time analysis of camera data is a challenge for both traditional and Machine Learning algorithms, since all computations in the MSL have to be performed on the robot itself.
This contribution shows that it is possible to process camera imagery in real-time using Machine Learning. It does this by presenting the current state of Machine Learning in MSL and providing two examples that won the Scientific and Technical Challenges at RoboCup 2021. Both examples focus on semantic detection of objects and humans in imagery. The Scientific Challenge winner presents how YOLOv5 can be used for object detection in the MSL. The Technical Challenge winner demonstrates how to improve interaction between robots and humans in soccer using OpenPose. This contributes towards the goal of RoboCup to arrive at robots that can beat the human soccer world champion by 2050.

Original language	English
Title of host publication	RoboCup 2021
Subtitle of host publication	Robot World Cup XXIV
Editors	Rachid Alami, Joydeep Biswas, Maya Cakmak, Oliver Obst
Place of Publication	Cham
Publisher	Springer
Chapter	27
Pages	327-339
Number of pages	13
ISBN (Electronic)	978-3-030-98682-7
ISBN (Print)	978-3-030-98681-0
DOIs	https://doi.org/10.1007/978-3-030-98682-7_27
Publication status	Published - 2022
Event	24th RoboCup International Symposium, RoboCup 2021 - Sydney, Australia Duration: 22 Jun 2021 → 28 Jun 2021 Conference number: 24 https://2021.robocup.org/

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13132 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	24th RoboCup International Symposium, RoboCup 2021
Abbreviated title	RoboCup 2021
Country/Territory	Australia
City	Sydney
Period	22/06/21 → 28/06/21
Other	Robot World Cup XXIV
Internet address	https://2021.robocup.org/

Keywords

RoboCup soccer
Middle size league
Machine learning
Object detection
People detection
Real-time

Access to Document

10.1007/978-3-030-98682-7_27

Cite this

Olthuis, J. J., van der Meer, N., Kempers, S. T., van Hoof, C. A., Beumer, R. M., Kuijpers, W. J. P., Kokkelmans, A. A., Houtman, W., van Eijck, J. J. F. J., Kon, J. J., Peijnenburg, A. T. A., & van de Molengraft, M. J. G. (2022). Vision-Based Machine Learning in Robot Soccer. In R. Alami, J. Biswas, M. Cakmak, & O. Obst (Eds.), RoboCup 2021: Robot World Cup XXIV (pp. 327-339). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13132 LNAI). Springer. https://doi.org/10.1007/978-3-030-98682-7_27

Olthuis, J.J. ; van der Meer, Noah ; Kempers, S.T. et al. / Vision-Based Machine Learning in Robot Soccer. RoboCup 2021: Robot World Cup XXIV. editor / Rachid Alami ; Joydeep Biswas ; Maya Cakmak ; Oliver Obst. Cham : Springer, 2022. pp. 327-339 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{10351986a4f6450db1e008c5a475d0d8,

title = "Vision-Based Machine Learning in Robot Soccer",

abstract = "Robots need to perceive their environment in order to properly interact with it. In the RoboCup Soccer Middle Size League (MSL) this happens primarily through cameras mounted on the robots. Machine Learning can be used to extract relevant features from camera imagery. The real-time analysis of camera data is a challenge for both traditional and Machine Learning algorithms, since all computations in the MSL have to be performed on the robot itself. This contribution shows that it is possible to process camera imagery in real-time using Machine Learning. It does this by presenting the current state of Machine Learning in MSL and providing two examples that won the Scientific and Technical Challenges at RoboCup 2021. Both examples focus on semantic detection of objects and humans in imagery. The Scientific Challenge winner presents how YOLOv5 can be used for object detection in the MSL. The Technical Challenge winner demonstrates how to improve interaction between robots and humans in soccer using OpenPose. This contributes towards the goal of RoboCup to arrive at robots that can beat the human soccer world champion by 2050.",

keywords = "RoboCup soccer, Middle size league, Machine learning, Object detection, People detection, Real-time",

author = "J.J. Olthuis and \{van der Meer\}, Noah and S.T. Kempers and \{van Hoof\}, C.A. and R.M. Beumer and W.J.P. Kuijpers and A.A. Kokkelmans and Wouter Houtman and \{van Eijck\}, J.J.F.J. and J.J. Kon and A.T.A. Peijnenburg and \{van de Molengraft\}, M.J.G.",

year = "2022",

doi = "10.1007/978-3-030-98682-7\_27",

language = "English",

isbn = "978-3-030-98681-0",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "327--339",

editor = "Rachid Alami and Joydeep Biswas and Maya Cakmak and Oliver Obst",

booktitle = "RoboCup 2021",

address = "Germany",

note = "24th RoboCup International Symposium, RoboCup 2021, RoboCup 2021 ; Conference date: 22-06-2021 Through 28-06-2021",

url = "https://2021.robocup.org/",

}

Olthuis, JJ, van der Meer, N, Kempers, ST, van Hoof, CA, Beumer, RM , Kuijpers, WJP, Kokkelmans, AA, Houtman, W, van Eijck, JJFJ, Kon, JJ, Peijnenburg, ATA & van de Molengraft, MJG 2022, Vision-Based Machine Learning in Robot Soccer. in R Alami, J Biswas, M Cakmak & O Obst (eds), RoboCup 2021: Robot World Cup XXIV. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13132 LNAI, Springer, Cham, pp. 327-339, 24th RoboCup International Symposium, RoboCup 2021, Sydney, New South Wales, Australia, 22/06/21. https://doi.org/10.1007/978-3-030-98682-7_27

Vision-Based Machine Learning in Robot Soccer. / Olthuis, J.J.; van der Meer, Noah; Kempers, S.T. et al.
RoboCup 2021: Robot World Cup XXIV. ed. / Rachid Alami; Joydeep Biswas; Maya Cakmak; Oliver Obst. Cham: Springer, 2022. p. 327-339 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13132 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Vision-Based Machine Learning in Robot Soccer

AU - Olthuis, J.J.

AU - van der Meer, Noah

AU - Kempers, S.T.

AU - van Hoof, C.A.

AU - Beumer, R.M.

AU - Kuijpers, W.J.P.

AU - Kokkelmans, A.A.

AU - Houtman, Wouter

AU - van Eijck, J.J.F.J.

AU - Kon, J.J.

AU - Peijnenburg, A.T.A.

AU - van de Molengraft, M.J.G.

N1 - Conference code: 24

PY - 2022

Y1 - 2022

N2 - Robots need to perceive their environment in order to properly interact with it. In the RoboCup Soccer Middle Size League (MSL) this happens primarily through cameras mounted on the robots. Machine Learning can be used to extract relevant features from camera imagery. The real-time analysis of camera data is a challenge for both traditional and Machine Learning algorithms, since all computations in the MSL have to be performed on the robot itself. This contribution shows that it is possible to process camera imagery in real-time using Machine Learning. It does this by presenting the current state of Machine Learning in MSL and providing two examples that won the Scientific and Technical Challenges at RoboCup 2021. Both examples focus on semantic detection of objects and humans in imagery. The Scientific Challenge winner presents how YOLOv5 can be used for object detection in the MSL. The Technical Challenge winner demonstrates how to improve interaction between robots and humans in soccer using OpenPose. This contributes towards the goal of RoboCup to arrive at robots that can beat the human soccer world champion by 2050.

AB - Robots need to perceive their environment in order to properly interact with it. In the RoboCup Soccer Middle Size League (MSL) this happens primarily through cameras mounted on the robots. Machine Learning can be used to extract relevant features from camera imagery. The real-time analysis of camera data is a challenge for both traditional and Machine Learning algorithms, since all computations in the MSL have to be performed on the robot itself. This contribution shows that it is possible to process camera imagery in real-time using Machine Learning. It does this by presenting the current state of Machine Learning in MSL and providing two examples that won the Scientific and Technical Challenges at RoboCup 2021. Both examples focus on semantic detection of objects and humans in imagery. The Scientific Challenge winner presents how YOLOv5 can be used for object detection in the MSL. The Technical Challenge winner demonstrates how to improve interaction between robots and humans in soccer using OpenPose. This contributes towards the goal of RoboCup to arrive at robots that can beat the human soccer world champion by 2050.

KW - RoboCup soccer

KW - Middle size league

KW - Machine learning

KW - Object detection

KW - People detection

KW - Real-time

UR - https://www.scopus.com/pages/publications/85127925864

U2 - 10.1007/978-3-030-98682-7_27

DO - 10.1007/978-3-030-98682-7_27

M3 - Conference contribution

SN - 978-3-030-98681-0

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 327

EP - 339

BT - RoboCup 2021

A2 - Alami, Rachid

A2 - Biswas, Joydeep

A2 - Cakmak, Maya

A2 - Obst, Oliver

PB - Springer

CY - Cham

T2 - 24th RoboCup International Symposium, RoboCup 2021

Y2 - 22 June 2021 through 28 June 2021

ER -

Olthuis JJ, van der Meer N, Kempers ST, van Hoof CA, Beumer RM , Kuijpers WJP et al. Vision-Based Machine Learning in Robot Soccer. In Alami R, Biswas J, Cakmak M, Obst O, editors, RoboCup 2021: Robot World Cup XXIV. Cham: Springer. 2022. p. 327-339. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-98682-7_27

Vision-Based Machine Learning in Robot Soccer

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this