Many-core acceleration for model predictive control systems

Satoshi Kawakami, Akihito Iwanaga, Koji Inoue

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

2 Citations (Scopus)

Abstract

This paper proposes a novel many-core execution strategy for real-time model predictive controls. The key idea is to exploit predicted input values, which are produced by the model predictive control itself, to speculatively solve an op- timal control problem. It is well known that control appli- cations are not suitable for multi- or many-core processors, because feedback-loop systems inherently stand on sequen- tial operations. Since the proposed scheme does not rely on conventional thread-/data-level parallelism, it can be easily applied to such control systems. An analytical evaluation using a real application demonstrates the potential of per- formance improvement achieved by the proposed speculative executions.

Original languageEnglish
Title of host publication1st International Workshop on Many-Core Embedded Systems, MES 2013 - In Conjunction with the 40th Annual IEEE/ACM International Symposium on Computer Architecture, ISCA 2013
Pages17-24
Number of pages8
DOIs
Publication statusPublished - 2013
Event1st International Workshop on Many-Core Embedded Systems, MES 2013, in Conjunction with the 40th Annual IEEE/ACM International Symposium on Computer Architecture, ISCA 2013 - Tel-Aviv, Israel
Duration: Jun 24 2013Jun 24 2013

Publication series

NameACM International Conference Proceeding Series

Other

Other1st International Workshop on Many-Core Embedded Systems, MES 2013, in Conjunction with the 40th Annual IEEE/ACM International Symposium on Computer Architecture, ISCA 2013
Country/TerritoryIsrael
CityTel-Aviv
Period6/24/136/24/13

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Many-core acceleration for model predictive control systems'. Together they form a unique fingerprint.

Cite this