Gated Recurrent Unit Based Short-Term Network for Robot Swarm Motion Forecasting

Belkacem Khaldi; Fouzi Harrou; Dairi Abdelkader; Ying Sun

doi:10.1109/RTSI55261.2022.9905085

Gated Recurrent Unit Based Short-Term Network for Robot Swarm Motion Forecasting

Belkacem Khaldi, Fouzi Harrou, Dairi Abdelkader, Ying Sun

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

Abstract

This work introduces a short term forecasting gated recurrent unites framework for swarm motion speed forecasting. This is motivated by the growing need of addressing such forecasting challenges in order to keep swarm robotic systems executing daily collective operations and accomplishing tasks more successfully in groups. The framework is based on the BiGRU model and its performances is compared to its base GRU model. The framework is built upon sensor measurements collected using a simulated swarm of e-puck robots performing a simple pattern formation task in a free/no-free obstacles scenarios. Findings show how accurate BiGRU is in forecasting swarm motion speed while compared to GRU.

Original language	English (US)
Title of host publication	2022 IEEE 7th Forum on Research and Technologies for Society and Industry Innovation (RTSI)
Publisher	IEEE
ISBN (Print)	978-1-6654-9740-4
DOIs	https://doi.org/10.1109/RTSI55261.2022.9905085
State	Published - Oct 4 2022

Access to Document

10.1109/RTSI55261.2022.9905085

Cite this

@inproceedings{9e4c07a80fc34c1fa66b1fdd61a04a96,

title = "Gated Recurrent Unit Based Short-Term Network for Robot Swarm Motion Forecasting",

abstract = "This work introduces a short term forecasting gated recurrent unites framework for swarm motion speed forecasting. This is motivated by the growing need of addressing such forecasting challenges in order to keep swarm robotic systems executing daily collective operations and accomplishing tasks more successfully in groups. The framework is based on the BiGRU model and its performances is compared to its base GRU model. The framework is built upon sensor measurements collected using a simulated swarm of e-puck robots performing a simple pattern formation task in a free/no-free obstacles scenarios. Findings show how accurate BiGRU is in forecasting swarm motion speed while compared to GRU.",

author = "Belkacem Khaldi and Fouzi Harrou and Dairi Abdelkader and Ying Sun",

note = "KAUST Repository Item: Exported on 2022-10-07 Acknowledged KAUST grant number(s): OSR-2019-CRG7-3800 Acknowledgements: This publication is based upon a collaboration work from the King Abdullah University of Science and Technology (KAUST) and the Laboratoire de Recherche en Informatique de Sidi Bel-Abbes (LabRI-SBA). The work is supported by KAUST Office of Sponsored Research (OSR) under Award No: OSR-2019-CRG7-3800.",

year = "2022",

month = oct,

day = "4",

doi = "10.1109/RTSI55261.2022.9905085",

language = "English (US)",

isbn = "978-1-6654-9740-4",

booktitle = "2022 IEEE 7th Forum on Research and Technologies for Society and Industry Innovation (RTSI)",

publisher = "IEEE",

}

TY - GEN

T1 - Gated Recurrent Unit Based Short-Term Network for Robot Swarm Motion Forecasting

AU - Khaldi, Belkacem

AU - Harrou, Fouzi

AU - Abdelkader, Dairi

AU - Sun, Ying

N1 - KAUST Repository Item: Exported on 2022-10-07 Acknowledged KAUST grant number(s): OSR-2019-CRG7-3800 Acknowledgements: This publication is based upon a collaboration work from the King Abdullah University of Science and Technology (KAUST) and the Laboratoire de Recherche en Informatique de Sidi Bel-Abbes (LabRI-SBA). The work is supported by KAUST Office of Sponsored Research (OSR) under Award No: OSR-2019-CRG7-3800.

PY - 2022/10/4

Y1 - 2022/10/4

N2 - This work introduces a short term forecasting gated recurrent unites framework for swarm motion speed forecasting. This is motivated by the growing need of addressing such forecasting challenges in order to keep swarm robotic systems executing daily collective operations and accomplishing tasks more successfully in groups. The framework is based on the BiGRU model and its performances is compared to its base GRU model. The framework is built upon sensor measurements collected using a simulated swarm of e-puck robots performing a simple pattern formation task in a free/no-free obstacles scenarios. Findings show how accurate BiGRU is in forecasting swarm motion speed while compared to GRU.

AB - This work introduces a short term forecasting gated recurrent unites framework for swarm motion speed forecasting. This is motivated by the growing need of addressing such forecasting challenges in order to keep swarm robotic systems executing daily collective operations and accomplishing tasks more successfully in groups. The framework is based on the BiGRU model and its performances is compared to its base GRU model. The framework is built upon sensor measurements collected using a simulated swarm of e-puck robots performing a simple pattern formation task in a free/no-free obstacles scenarios. Findings show how accurate BiGRU is in forecasting swarm motion speed while compared to GRU.

UR - http://hdl.handle.net/10754/682247

UR - https://ieeexplore.ieee.org/document/9905085/

U2 - 10.1109/RTSI55261.2022.9905085

DO - 10.1109/RTSI55261.2022.9905085

M3 - Conference contribution

SN - 978-1-6654-9740-4

BT - 2022 IEEE 7th Forum on Research and Technologies for Society and Industry Innovation (RTSI)

PB - IEEE

ER -

Gated Recurrent Unit Based Short-Term Network for Robot Swarm Motion Forecasting

Abstract

Access to Document

Other files and links

Fingerprint

Cite this