An Approach to Self- Learning Multicore Reconfiguration Management Applied on Robotic Vision

Walter Stechele; Jan Hartmann; Erik Maehle

doi:10.1109/DASIP.2011.6136882

An Approach to Self- Learning Multicore Reconfiguration Management Applied on Robotic Vision

Walter Stechele, Jan Hartmann, Erik Maehle

Institute of Computer Engineering

Abstract

Robotic Vision combined with real-time control imposes challenging requirements on embedded computing nodes in robots, exhibiting strong variations in computational load due to dynamically changing activity profiles. Reconfigurable Multiprocessor System-on-Chip offers a solution by efficiently handling the robot's resources, but reconfiguration management seems challenging. The goal of this paper is to present first ideas on self-learning reconfiguration management for Reco nfigurable multicore computing nodes with dynamic reconfiguration of soft-core CPUs and HW accelerators, to support dynamically changing activity profiles in Robotic Vision scenarios.

Original language	English
Title of host publication	Proceedings of the 2011 Conference on Design & Architectures for Signal & Image Processing (DASIP)
Number of pages	6
Publisher	IEEE
Publication date	01.12.2011
Pages	217-222
Article number	6136882
ISBN (Print)	978-1-4577-0620-2
ISBN (Electronic)	978-1-4577-0621-9, 978-1-4577-0619-6
DOIs	https://doi.org/10.1109/DASIP.2011.6136882
Publication status	Published - 01.12.2011
Event	2011 Conference on Design and Architectures for Signal and Image Processing - Tampere, Finland Duration: 02.11.2011 → 04.11.2011 Conference number: 88850

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/DASIP.2011.6136882

Cite this

@inproceedings{2cddc6763141475a9263a12d76f706bb,

title = "An Approach to Self- Learning Multicore Reconfiguration Management Applied on Robotic Vision",

abstract = "Robotic Vision combined with real-time control imposes challenging requirements on embedded computing nodes in robots, exhibiting strong variations in computational load due to dynamically changing activity profiles. Reconfigurable Multiprocessor System-on-Chip offers a solution by efficiently handling the robot's resources, but reconfiguration management seems challenging. The goal of this paper is to present first ideas on self-learning reconfiguration management for Reco nfigurable multicore computing nodes with dynamic reconfiguration of soft-core CPUs and HW accelerators, to support dynamically changing activity profiles in Robotic Vision scenarios.",

author = "Walter Stechele and Jan Hartmann and Erik Maehle",

year = "2011",

month = dec,

day = "1",

doi = "10.1109/DASIP.2011.6136882",

language = "English",

isbn = "978-1-4577-0620-2",

series = "IEEE Proceedings",

publisher = "IEEE",

pages = " 217--222",

booktitle = "Proceedings of the 2011 Conference on Design \& Architectures for Signal \& Image Processing (DASIP)",

note = "2011 Conference on Design and Architectures for Signal and Image Processing <br/>, DASIP 2011 ; Conference date: 02-11-2011 Through 04-11-2011",

}

Stechele, W, Hartmann, J & Maehle, E 2011, An Approach to Self- Learning Multicore Reconfiguration Management Applied on Robotic Vision. in Proceedings of the 2011 Conference on Design & Architectures for Signal & Image Processing (DASIP)., 6136882, IEEE Proceedings, IEEE, pp. 217-222, 2011 Conference on Design and Architectures for Signal and Image Processing
, Tampere, Finland, 02.11.11. https://doi.org/10.1109/DASIP.2011.6136882

An Approach to Self- Learning Multicore Reconfiguration Management Applied on Robotic Vision. / Stechele, Walter; Hartmann, Jan; Maehle, Erik.
Proceedings of the 2011 Conference on Design & Architectures for Signal & Image Processing (DASIP). IEEE, 2011. p. 217-222 6136882 (IEEE Proceedings).

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - An Approach to Self- Learning Multicore Reconfiguration Management Applied on Robotic Vision

AU - Stechele, Walter

AU - Hartmann, Jan

AU - Maehle, Erik

N1 - Conference code: 88850

PY - 2011/12/1

Y1 - 2011/12/1

N2 - Robotic Vision combined with real-time control imposes challenging requirements on embedded computing nodes in robots, exhibiting strong variations in computational load due to dynamically changing activity profiles. Reconfigurable Multiprocessor System-on-Chip offers a solution by efficiently handling the robot's resources, but reconfiguration management seems challenging. The goal of this paper is to present first ideas on self-learning reconfiguration management for Reco nfigurable multicore computing nodes with dynamic reconfiguration of soft-core CPUs and HW accelerators, to support dynamically changing activity profiles in Robotic Vision scenarios.

AB - Robotic Vision combined with real-time control imposes challenging requirements on embedded computing nodes in robots, exhibiting strong variations in computational load due to dynamically changing activity profiles. Reconfigurable Multiprocessor System-on-Chip offers a solution by efficiently handling the robot's resources, but reconfiguration management seems challenging. The goal of this paper is to present first ideas on self-learning reconfiguration management for Reco nfigurable multicore computing nodes with dynamic reconfiguration of soft-core CPUs and HW accelerators, to support dynamically changing activity profiles in Robotic Vision scenarios.

UR - https://www.scopus.com/pages/publications/84857695709

U2 - 10.1109/DASIP.2011.6136882

DO - 10.1109/DASIP.2011.6136882

M3 - Conference contribution

SN - 978-1-4577-0620-2

T3 - IEEE Proceedings

SP - 217

EP - 222

BT - Proceedings of the 2011 Conference on Design & Architectures for Signal & Image Processing (DASIP)

PB - IEEE

T2 - 2011 Conference on Design and Architectures for Signal and Image Processing <br/>

Y2 - 2 November 2011 through 4 November 2011

ER -

An Approach to Self- Learning Multicore Reconfiguration Management Applied on Robotic Vision

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this