Approximate LASSO model predictive control for resource constrained systems

Yun Wu; Joao F.C. Mota; Andrew M. Wallace

doi:10.1109/SSPD47486.2020.9272000

Approximate LASSO model predictive control for resource constrained systems

Yun Wu
, Joao F.C. Mota
, Andrew M. Wallace

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Citations (Scopus)

51 Downloads (Pure)

Abstract

LASSO MPC is a popular method for solving optimal control problems within a receding horizon. It is, however, challenging to deploy LASSO MPC on resource constrained systems, such as embedded platforms, due to the intensive memory usage and computational cost as the horizon length is extended. By exploiting a reduced precision, approximation technique applied to Proximal Gradient Descent (PGD), we demonstrate an implementation on a resource constrained, reconfigurable device, such as a Field Programmable Gate Array (FPGA). Our experiments show equivalent performance to a high-precision optimisation solver, but with significant improvements to both logic cost and memory bandwidth, up to 60% and 80% reduction respectively, with up to 70% power savings.

Original language	English
Title of host publication	2020 Sensor Signal Processing for Defence Conference (SSPD): Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728138107
ISBN (Print)	9781728138114
DOIs	https://doi.org/10.1109/SSPD47486.2020.9272000
Publication status	Published - 30 Nov 2020
Externally published	Yes
Event	9th Sensor Signal Processing for Defence Conference, SSPD 2020 - Edinburgh, United Kingdom Duration: 15 Sept 2020 → 16 Sept 2020

Publication series

Name	Sensor Signal Processing for Defence Conference (SSPD): Proceedings

Conference

Conference	9th Sensor Signal Processing for Defence Conference, SSPD 2020
Country/Territory	United Kingdom
City	Edinburgh
Period	15/09/2020 → 16/09/2020

Bibliographical note

Funding Information:
Work supported by EPSRC Grant number EP/S000631/1 and the MOD University Defence Research Collaboration (UDRC) in Signal Processing.

Publisher Copyright:
© 2020 IEEE.

Keywords

Approximate Computing
FPGA
LASSO
MPC
Proximal Gradient Descent

ASJC Scopus subject areas

Artificial Intelligence
Signal Processing

Access to Document

10.1109/SSPD47486.2020.9272000Licence: Unspecified

Approximate LASSO model predictive control for resource constrained systems
Copyright 2020, IEEE. This work is made available online in accordance with the publisher’s policies. Please refer to any applicable terms of use of the publisher
Accepted author manuscript, 403 KBLicence: Unspecified

Cite this

Wu, Y., Mota, J. F. C., & Wallace, A. M. (2020). Approximate LASSO model predictive control for resource constrained systems. In 2020 Sensor Signal Processing for Defence Conference (SSPD): Proceedings Article 9272000 ( Sensor Signal Processing for Defence Conference (SSPD): Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SSPD47486.2020.9272000

@inproceedings{28460517f5104ea783b695dfa2f8e3d0,

title = "Approximate LASSO model predictive control for resource constrained systems",

abstract = "LASSO MPC is a popular method for solving optimal control problems within a receding horizon. It is, however, challenging to deploy LASSO MPC on resource constrained systems, such as embedded platforms, due to the intensive memory usage and computational cost as the horizon length is extended. By exploiting a reduced precision, approximation technique applied to Proximal Gradient Descent (PGD), we demonstrate an implementation on a resource constrained, reconfigurable device, such as a Field Programmable Gate Array (FPGA). Our experiments show equivalent performance to a high-precision optimisation solver, but with significant improvements to both logic cost and memory bandwidth, up to 60\% and 80\% reduction respectively, with up to 70\% power savings.",

keywords = "Approximate Computing, FPGA, LASSO, MPC, Proximal Gradient Descent",

author = "Yun Wu and Mota, \{Joao F.C.\} and Wallace, \{Andrew M.\}",

note = "Funding Information: Work supported by EPSRC Grant number EP/S000631/1 and the MOD University Defence Research Collaboration (UDRC) in Signal Processing. Publisher Copyright: {\textcopyright} 2020 IEEE.; 9th Sensor Signal Processing for Defence Conference, SSPD 2020 ; Conference date: 15-09-2020 Through 16-09-2020",

year = "2020",

month = nov,

day = "30",

doi = "10.1109/SSPD47486.2020.9272000",

language = "English",

isbn = "9781728138114",

series = " Sensor Signal Processing for Defence Conference (SSPD): Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2020 Sensor Signal Processing for Defence Conference (SSPD): Proceedings",

address = "United States",

}

Wu, Y, Mota, JFC & Wallace, AM 2020, Approximate LASSO model predictive control for resource constrained systems. in 2020 Sensor Signal Processing for Defence Conference (SSPD): Proceedings., 9272000, Sensor Signal Processing for Defence Conference (SSPD): Proceedings, Institute of Electrical and Electronics Engineers Inc., 9th Sensor Signal Processing for Defence Conference, SSPD 2020, Edinburgh, United Kingdom, 15/09/2020. https://doi.org/10.1109/SSPD47486.2020.9272000

Approximate LASSO model predictive control for resource constrained systems. / Wu, Yun; Mota, Joao F.C.; Wallace, Andrew M.
2020 Sensor Signal Processing for Defence Conference (SSPD): Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. 9272000 ( Sensor Signal Processing for Defence Conference (SSPD): Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Approximate LASSO model predictive control for resource constrained systems

AU - Wu, Yun

AU - Mota, Joao F.C.

AU - Wallace, Andrew M.

PY - 2020/11/30

Y1 - 2020/11/30

N2 - LASSO MPC is a popular method for solving optimal control problems within a receding horizon. It is, however, challenging to deploy LASSO MPC on resource constrained systems, such as embedded platforms, due to the intensive memory usage and computational cost as the horizon length is extended. By exploiting a reduced precision, approximation technique applied to Proximal Gradient Descent (PGD), we demonstrate an implementation on a resource constrained, reconfigurable device, such as a Field Programmable Gate Array (FPGA). Our experiments show equivalent performance to a high-precision optimisation solver, but with significant improvements to both logic cost and memory bandwidth, up to 60% and 80% reduction respectively, with up to 70% power savings.

AB - LASSO MPC is a popular method for solving optimal control problems within a receding horizon. It is, however, challenging to deploy LASSO MPC on resource constrained systems, such as embedded platforms, due to the intensive memory usage and computational cost as the horizon length is extended. By exploiting a reduced precision, approximation technique applied to Proximal Gradient Descent (PGD), we demonstrate an implementation on a resource constrained, reconfigurable device, such as a Field Programmable Gate Array (FPGA). Our experiments show equivalent performance to a high-precision optimisation solver, but with significant improvements to both logic cost and memory bandwidth, up to 60% and 80% reduction respectively, with up to 70% power savings.

KW - Approximate Computing

KW - FPGA

KW - LASSO

KW - MPC

KW - Proximal Gradient Descent

U2 - 10.1109/SSPD47486.2020.9272000

DO - 10.1109/SSPD47486.2020.9272000

M3 - Conference contribution

AN - SCOPUS:85098582359

SN - 9781728138114

T3 - Sensor Signal Processing for Defence Conference (SSPD): Proceedings

BT - 2020 Sensor Signal Processing for Defence Conference (SSPD): Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 9th Sensor Signal Processing for Defence Conference, SSPD 2020

Y2 - 15 September 2020 through 16 September 2020

ER -

Approximate LASSO model predictive control for resource constrained systems

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this