VLSI platform for real-world intelligent integrated systems based on algorithm selection

Martin Lukac, Michitaka Kameyama, Yoshichika Fujioka

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A real-world intelligent system consists of three basic modules: environment recognition, prediction (or estimation), and behavior planning. To obtain high quality results in these modules, high speed processing and real time adaptability on a case by case basis are required. In each of the above mentioned modules, many different algorithms and algorithms networks exists and provide various performances on a case by case basis. Thus, a mechanism that for any of the three computational stages selects the best possible algorithm is required. We propose a platform based on the algorithm selection approach to the problem of natural image understanding. This selection mechanism is based on machine learning; a bottom-up algorithm selection from real-world image features and a top-down algorithm selection using information obtained from a high level symbolic world description and algorithm suitability. To accommodate the highspeed processing requirements, the high-frequency of real-time reconfiguration and a low-cost of implementation, we are using present a novel dynamic reconfigurable VLSI processor for real-time adaptation of the algorithm selection. The new architecture includes a fine-grain Digital Reconfigurable Processor, a distributed configuration memory to solve the data transfer bottleneck and an intra-chip packet routing scheme to reduce the size of the configuration memory.

Original languageEnglish
Title of host publicationProc. of the IADIS Int. Conf. Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, Proc. of the IADIS Int. Conf. Theory and Practice in Modern Computing 2013, TPMC 2013
Pages27-34
Number of pages8
Publication statusPublished - 2013
Externally publishedYes
EventIADIS International Conference Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, IADIS International Conference TPMC 2013, Part of the IADIS Multi Conference on Computer Science and Information Systems 2013, MCCSIS 2013 - Prague, Czech Republic
Duration: Jul 22 2013Jul 24 2013

Other

OtherIADIS International Conference Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, IADIS International Conference TPMC 2013, Part of the IADIS Multi Conference on Computer Science and Information Systems 2013, MCCSIS 2013
CountryCzech Republic
CityPrague
Period7/22/137/24/13

Fingerprint

Image understanding
Data storage equipment
Intelligent systems
Data transfer
Processing
Learning systems
Planning
Costs

Keywords

  • Algorithm selection
  • Memory reduction
  • Natural image processing
  • Reconfigurable VLSI

ASJC Scopus subject areas

  • Information Systems

Cite this

Lukac, M., Kameyama, M., & Fujioka, Y. (2013). VLSI platform for real-world intelligent integrated systems based on algorithm selection. In Proc. of the IADIS Int. Conf. Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, Proc. of the IADIS Int. Conf. Theory and Practice in Modern Computing 2013, TPMC 2013 (pp. 27-34)

VLSI platform for real-world intelligent integrated systems based on algorithm selection. / Lukac, Martin; Kameyama, Michitaka; Fujioka, Yoshichika.

Proc. of the IADIS Int. Conf. Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, Proc. of the IADIS Int. Conf. Theory and Practice in Modern Computing 2013, TPMC 2013. 2013. p. 27-34.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lukac, M, Kameyama, M & Fujioka, Y 2013, VLSI platform for real-world intelligent integrated systems based on algorithm selection. in Proc. of the IADIS Int. Conf. Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, Proc. of the IADIS Int. Conf. Theory and Practice in Modern Computing 2013, TPMC 2013. pp. 27-34, IADIS International Conference Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, IADIS International Conference TPMC 2013, Part of the IADIS Multi Conference on Computer Science and Information Systems 2013, MCCSIS 2013, Prague, Czech Republic, 7/22/13.
Lukac M, Kameyama M, Fujioka Y. VLSI platform for real-world intelligent integrated systems based on algorithm selection. In Proc. of the IADIS Int. Conf. Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, Proc. of the IADIS Int. Conf. Theory and Practice in Modern Computing 2013, TPMC 2013. 2013. p. 27-34
Lukac, Martin ; Kameyama, Michitaka ; Fujioka, Yoshichika. / VLSI platform for real-world intelligent integrated systems based on algorithm selection. Proc. of the IADIS Int. Conf. Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, Proc. of the IADIS Int. Conf. Theory and Practice in Modern Computing 2013, TPMC 2013. 2013. pp. 27-34
@inproceedings{56c484a2c07e47c485e5058791f00f32,
title = "VLSI platform for real-world intelligent integrated systems based on algorithm selection",
abstract = "A real-world intelligent system consists of three basic modules: environment recognition, prediction (or estimation), and behavior planning. To obtain high quality results in these modules, high speed processing and real time adaptability on a case by case basis are required. In each of the above mentioned modules, many different algorithms and algorithms networks exists and provide various performances on a case by case basis. Thus, a mechanism that for any of the three computational stages selects the best possible algorithm is required. We propose a platform based on the algorithm selection approach to the problem of natural image understanding. This selection mechanism is based on machine learning; a bottom-up algorithm selection from real-world image features and a top-down algorithm selection using information obtained from a high level symbolic world description and algorithm suitability. To accommodate the highspeed processing requirements, the high-frequency of real-time reconfiguration and a low-cost of implementation, we are using present a novel dynamic reconfigurable VLSI processor for real-time adaptation of the algorithm selection. The new architecture includes a fine-grain Digital Reconfigurable Processor, a distributed configuration memory to solve the data transfer bottleneck and an intra-chip packet routing scheme to reduce the size of the configuration memory.",
keywords = "Algorithm selection, Memory reduction, Natural image processing, Reconfigurable VLSI",
author = "Martin Lukac and Michitaka Kameyama and Yoshichika Fujioka",
year = "2013",
language = "English",
isbn = "9789728939946",
pages = "27--34",
booktitle = "Proc. of the IADIS Int. Conf. Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, Proc. of the IADIS Int. Conf. Theory and Practice in Modern Computing 2013, TPMC 2013",

}

TY - GEN

T1 - VLSI platform for real-world intelligent integrated systems based on algorithm selection

AU - Lukac, Martin

AU - Kameyama, Michitaka

AU - Fujioka, Yoshichika

PY - 2013

Y1 - 2013

N2 - A real-world intelligent system consists of three basic modules: environment recognition, prediction (or estimation), and behavior planning. To obtain high quality results in these modules, high speed processing and real time adaptability on a case by case basis are required. In each of the above mentioned modules, many different algorithms and algorithms networks exists and provide various performances on a case by case basis. Thus, a mechanism that for any of the three computational stages selects the best possible algorithm is required. We propose a platform based on the algorithm selection approach to the problem of natural image understanding. This selection mechanism is based on machine learning; a bottom-up algorithm selection from real-world image features and a top-down algorithm selection using information obtained from a high level symbolic world description and algorithm suitability. To accommodate the highspeed processing requirements, the high-frequency of real-time reconfiguration and a low-cost of implementation, we are using present a novel dynamic reconfigurable VLSI processor for real-time adaptation of the algorithm selection. The new architecture includes a fine-grain Digital Reconfigurable Processor, a distributed configuration memory to solve the data transfer bottleneck and an intra-chip packet routing scheme to reduce the size of the configuration memory.

AB - A real-world intelligent system consists of three basic modules: environment recognition, prediction (or estimation), and behavior planning. To obtain high quality results in these modules, high speed processing and real time adaptability on a case by case basis are required. In each of the above mentioned modules, many different algorithms and algorithms networks exists and provide various performances on a case by case basis. Thus, a mechanism that for any of the three computational stages selects the best possible algorithm is required. We propose a platform based on the algorithm selection approach to the problem of natural image understanding. This selection mechanism is based on machine learning; a bottom-up algorithm selection from real-world image features and a top-down algorithm selection using information obtained from a high level symbolic world description and algorithm suitability. To accommodate the highspeed processing requirements, the high-frequency of real-time reconfiguration and a low-cost of implementation, we are using present a novel dynamic reconfigurable VLSI processor for real-time adaptation of the algorithm selection. The new architecture includes a fine-grain Digital Reconfigurable Processor, a distributed configuration memory to solve the data transfer bottleneck and an intra-chip packet routing scheme to reduce the size of the configuration memory.

KW - Algorithm selection

KW - Memory reduction

KW - Natural image processing

KW - Reconfigurable VLSI

UR - http://www.scopus.com/inward/record.url?scp=84886903365&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84886903365&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9789728939946

SP - 27

EP - 34

BT - Proc. of the IADIS Int. Conf. Information Systems Post-Implementation and Change Management 2013, ISPCM 2013, Proc. of the IADIS Int. Conf. Theory and Practice in Modern Computing 2013, TPMC 2013

ER -