Model-Driven Optimization of Data-Adaptable Embedded Systems

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

Abstract

Complex sensing and decision applications such as object tracking and classification, video surveillance, unmanned aerial vehicle flight decisions, and others operate on vast data streams with dynamic characteristics. As the availability and quality of the sensed data changes, the underlying models and decision algorithms should continually adapt in order to meet desired high-level requirements. Due to the complexity of such dynamic data-driven systems, traditional design time techniques are often incapable of producing a solution that remains optimal in the face of dynamically changing data, algorithms, and even availability of computational resources. To assist developers of these systems, we present a modeling and optimization methodology that enables developers to capture application task flows and data sources, define associated quality metrics with data types, specify each algorithm's data and quality requirements, and define a data quality estimation framework to optimize the application at runtime. We demonstrate each facet of the modeling and optimization process via a video-based vehicle tracking and collision avoidance application, and show how such an approach results in efficient design space exploration when selecting the optimal set of algorithm modalities. When searching for an application configuration within 1% to 5% of optimal, our model-guided approach can achieve speedups of up to 9.3X versus a standard genetic algorithm and speedups of up to 80X relative to a brute force algorithm.

Original languageEnglish (US)
Title of host publicationProceedings - 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016
PublisherIEEE Computer Society
Pages293-302
Number of pages10
Volume1
ISBN (Electronic)9781467388450
DOIs
StatePublished - Aug 24 2016
Event2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016 - Atlanta, United States
Duration: Jun 10 2016Jun 14 2016

Other

Other2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016
CountryUnited States
CityAtlanta
Period6/10/166/14/16

Fingerprint

Embedded systems
Availability
Collision avoidance
Unmanned aerial vehicles (UAV)
Genetic algorithms

Keywords

  • design space exploration
  • dynamic data-driven systems
  • dynamic optimization
  • Software modeling

ASJC Scopus subject areas

  • Software

Cite this

Lizarraga, A., Lysecky, R. L., & Sprinkle, J. (2016). Model-Driven Optimization of Data-Adaptable Embedded Systems. In Proceedings - 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016 (Vol. 1, pp. 293-302). [7552025] IEEE Computer Society. https://doi.org/10.1109/COMPSAC.2016.156

Model-Driven Optimization of Data-Adaptable Embedded Systems. / Lizarraga, Adrian; Lysecky, Roman L; Sprinkle, Jonathan.

Proceedings - 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016. Vol. 1 IEEE Computer Society, 2016. p. 293-302 7552025.

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

Lizarraga, A, Lysecky, RL & Sprinkle, J 2016, Model-Driven Optimization of Data-Adaptable Embedded Systems. in Proceedings - 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016. vol. 1, 7552025, IEEE Computer Society, pp. 293-302, 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016, Atlanta, United States, 6/10/16. https://doi.org/10.1109/COMPSAC.2016.156
Lizarraga A, Lysecky RL, Sprinkle J. Model-Driven Optimization of Data-Adaptable Embedded Systems. In Proceedings - 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016. Vol. 1. IEEE Computer Society. 2016. p. 293-302. 7552025 https://doi.org/10.1109/COMPSAC.2016.156
Lizarraga, Adrian ; Lysecky, Roman L ; Sprinkle, Jonathan. / Model-Driven Optimization of Data-Adaptable Embedded Systems. Proceedings - 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016. Vol. 1 IEEE Computer Society, 2016. pp. 293-302
@inproceedings{facca91428604943ac3449460988d5b1,
title = "Model-Driven Optimization of Data-Adaptable Embedded Systems",
abstract = "Complex sensing and decision applications such as object tracking and classification, video surveillance, unmanned aerial vehicle flight decisions, and others operate on vast data streams with dynamic characteristics. As the availability and quality of the sensed data changes, the underlying models and decision algorithms should continually adapt in order to meet desired high-level requirements. Due to the complexity of such dynamic data-driven systems, traditional design time techniques are often incapable of producing a solution that remains optimal in the face of dynamically changing data, algorithms, and even availability of computational resources. To assist developers of these systems, we present a modeling and optimization methodology that enables developers to capture application task flows and data sources, define associated quality metrics with data types, specify each algorithm's data and quality requirements, and define a data quality estimation framework to optimize the application at runtime. We demonstrate each facet of the modeling and optimization process via a video-based vehicle tracking and collision avoidance application, and show how such an approach results in efficient design space exploration when selecting the optimal set of algorithm modalities. When searching for an application configuration within 1{\%} to 5{\%} of optimal, our model-guided approach can achieve speedups of up to 9.3X versus a standard genetic algorithm and speedups of up to 80X relative to a brute force algorithm.",
keywords = "design space exploration, dynamic data-driven systems, dynamic optimization, Software modeling",
author = "Adrian Lizarraga and Lysecky, {Roman L} and Jonathan Sprinkle",
year = "2016",
month = "8",
day = "24",
doi = "10.1109/COMPSAC.2016.156",
language = "English (US)",
volume = "1",
pages = "293--302",
booktitle = "Proceedings - 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Model-Driven Optimization of Data-Adaptable Embedded Systems

AU - Lizarraga, Adrian

AU - Lysecky, Roman L

AU - Sprinkle, Jonathan

PY - 2016/8/24

Y1 - 2016/8/24

N2 - Complex sensing and decision applications such as object tracking and classification, video surveillance, unmanned aerial vehicle flight decisions, and others operate on vast data streams with dynamic characteristics. As the availability and quality of the sensed data changes, the underlying models and decision algorithms should continually adapt in order to meet desired high-level requirements. Due to the complexity of such dynamic data-driven systems, traditional design time techniques are often incapable of producing a solution that remains optimal in the face of dynamically changing data, algorithms, and even availability of computational resources. To assist developers of these systems, we present a modeling and optimization methodology that enables developers to capture application task flows and data sources, define associated quality metrics with data types, specify each algorithm's data and quality requirements, and define a data quality estimation framework to optimize the application at runtime. We demonstrate each facet of the modeling and optimization process via a video-based vehicle tracking and collision avoidance application, and show how such an approach results in efficient design space exploration when selecting the optimal set of algorithm modalities. When searching for an application configuration within 1% to 5% of optimal, our model-guided approach can achieve speedups of up to 9.3X versus a standard genetic algorithm and speedups of up to 80X relative to a brute force algorithm.

AB - Complex sensing and decision applications such as object tracking and classification, video surveillance, unmanned aerial vehicle flight decisions, and others operate on vast data streams with dynamic characteristics. As the availability and quality of the sensed data changes, the underlying models and decision algorithms should continually adapt in order to meet desired high-level requirements. Due to the complexity of such dynamic data-driven systems, traditional design time techniques are often incapable of producing a solution that remains optimal in the face of dynamically changing data, algorithms, and even availability of computational resources. To assist developers of these systems, we present a modeling and optimization methodology that enables developers to capture application task flows and data sources, define associated quality metrics with data types, specify each algorithm's data and quality requirements, and define a data quality estimation framework to optimize the application at runtime. We demonstrate each facet of the modeling and optimization process via a video-based vehicle tracking and collision avoidance application, and show how such an approach results in efficient design space exploration when selecting the optimal set of algorithm modalities. When searching for an application configuration within 1% to 5% of optimal, our model-guided approach can achieve speedups of up to 9.3X versus a standard genetic algorithm and speedups of up to 80X relative to a brute force algorithm.

KW - design space exploration

KW - dynamic data-driven systems

KW - dynamic optimization

KW - Software modeling

UR - http://www.scopus.com/inward/record.url?scp=84987948214&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84987948214&partnerID=8YFLogxK

U2 - 10.1109/COMPSAC.2016.156

DO - 10.1109/COMPSAC.2016.156

M3 - Conference contribution

AN - SCOPUS:84987948214

VL - 1

SP - 293

EP - 302

BT - Proceedings - 2016 IEEE 40th Annual Computer Software and Applications Conference, COMPSAC 2016

PB - IEEE Computer Society

ER -