Edge-accelerated UAV operations: a case study of open source solutions

Julio Diez-Tomillo; Jose M. Alcaraz-Calero; Qi Wang

doi:10.1109/IWCMC61514.2024.10592445

Edge-accelerated UAV operations: a case study of open source solutions

Julio Diez-Tomillo^*
, Jose M. Alcaraz-Calero
, Qi Wang

^*Corresponding author for this work

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

110 Downloads (Pure)

Abstract

This study explores the execution of AI algorithms on open Unmanned Aerial Vehicles (UAVs) equipped with Beagle-Bone AI-64 (BBAI-64) boards, comparing their performance to high-performance computers equipped with GPUs. Key factors are evaluated, such as inference time, end-to-end processing time, CPU usage, or temperature on the board. Furthermore, this study presents the development of an open UAV platform based on an open-source flight controller (Durandal) executing an open-source autopilot (ArduPilot). This platform facilitates the integration of various sensors or cameras, regardless of brand or communication protocol. The study’s key findings show that the BBAI-64 offers advantages for smaller Artificial Intelligence (AI) models, and achieving comparable performance for larger models with high-performance computers. This work contributes to optimising AI execution on UAVs and supporting the development of versatile, sensor-agnostic open-source UAVs.

Original language	English
Title of host publication	Proceedings of the 20th International Wireless Communications & Mobile Computing Conference
Publisher	IEEE
Number of pages	7
DOIs	https://doi.org/10.1109/IWCMC61514.2024.10592445
Publication status	Published - 17 Jul 2024
Event	The 20th International Wireless Communications & Mobile Computing Conference: Green and Intelligent Communications - Adams Beach Hotel, Ayia Napa, Cyprus Duration: 27 May 2024 → 31 May 2024 https://iwcmc.org/2024/index.php

Conference

Conference	The 20th International Wireless Communications & Mobile Computing Conference
Abbreviated title	IWCMC 2024
Country/Territory	Cyprus
City	Ayia Napa
Period	27/05/24 → 31/05/24
Internet address	https://iwcmc.org/2024/index.php

Keywords

edge processing
BeagleBone AI-64
YOLOv5
UAV
open-source

Access to Document

10.1109/IWCMC61514.2024.10592445

2024 03 27 Diez-Tomillo et al UAV accepted - for PureAccepted author manuscript, 1.5 MBLicence: CC BY 4.0

Cite this

@inproceedings{d70ff2b5ea3e414b983e2c102fedcd0e,

title = "Edge-accelerated UAV operations: a case study of open source solutions",

abstract = "This study explores the execution of AI algorithms on open Unmanned Aerial Vehicles (UAVs) equipped with Beagle-Bone AI-64 (BBAI-64) boards, comparing their performance to high-performance computers equipped with GPUs. Key factors are evaluated, such as inference time, end-to-end processing time, CPU usage, or temperature on the board. Furthermore, this study presents the development of an open UAV platform based on an open-source flight controller (Durandal) executing an open-source autopilot (ArduPilot). This platform facilitates the integration of various sensors or cameras, regardless of brand or communication protocol. The study{\textquoteright}s key findings show that the BBAI-64 offers advantages for smaller Artificial Intelligence (AI) models, and achieving comparable performance for larger models with high-performance computers. This work contributes to optimising AI execution on UAVs and supporting the development of versatile, sensor-agnostic open-source UAVs.",

keywords = "edge processing, BeagleBone AI-64, YOLOv5, UAV, open-source",

author = "Julio Diez-Tomillo and Alcaraz-Calero, \{Jose M.\} and Qi Wang",

year = "2024",

month = jul,

day = "17",

doi = "10.1109/IWCMC61514.2024.10592445",

language = "English",

booktitle = "Proceedings of the 20th International Wireless Communications \& Mobile Computing Conference",

publisher = "IEEE",

address = "United States",

note = "The 20th International Wireless Communications \& Mobile Computing Conference : Green and Intelligent Communications, IWCMC 2024 ; Conference date: 27-05-2024 Through 31-05-2024",

url = "https://iwcmc.org/2024/index.php",

}

Diez-Tomillo, J, Alcaraz-Calero, JM & Wang, Q 2024, Edge-accelerated UAV operations: a case study of open source solutions. in Proceedings of the 20th International Wireless Communications & Mobile Computing Conference. IEEE, The 20th International Wireless Communications & Mobile Computing Conference, Ayia Napa, Cyprus, 27/05/24. https://doi.org/10.1109/IWCMC61514.2024.10592445

TY - GEN

T1 - Edge-accelerated UAV operations

T2 - The 20th International Wireless Communications & Mobile Computing Conference

AU - Diez-Tomillo, Julio

AU - Alcaraz-Calero, Jose M.

AU - Wang, Qi

PY - 2024/7/17

Y1 - 2024/7/17

N2 - This study explores the execution of AI algorithms on open Unmanned Aerial Vehicles (UAVs) equipped with Beagle-Bone AI-64 (BBAI-64) boards, comparing their performance to high-performance computers equipped with GPUs. Key factors are evaluated, such as inference time, end-to-end processing time, CPU usage, or temperature on the board. Furthermore, this study presents the development of an open UAV platform based on an open-source flight controller (Durandal) executing an open-source autopilot (ArduPilot). This platform facilitates the integration of various sensors or cameras, regardless of brand or communication protocol. The study’s key findings show that the BBAI-64 offers advantages for smaller Artificial Intelligence (AI) models, and achieving comparable performance for larger models with high-performance computers. This work contributes to optimising AI execution on UAVs and supporting the development of versatile, sensor-agnostic open-source UAVs.

AB - This study explores the execution of AI algorithms on open Unmanned Aerial Vehicles (UAVs) equipped with Beagle-Bone AI-64 (BBAI-64) boards, comparing their performance to high-performance computers equipped with GPUs. Key factors are evaluated, such as inference time, end-to-end processing time, CPU usage, or temperature on the board. Furthermore, this study presents the development of an open UAV platform based on an open-source flight controller (Durandal) executing an open-source autopilot (ArduPilot). This platform facilitates the integration of various sensors or cameras, regardless of brand or communication protocol. The study’s key findings show that the BBAI-64 offers advantages for smaller Artificial Intelligence (AI) models, and achieving comparable performance for larger models with high-performance computers. This work contributes to optimising AI execution on UAVs and supporting the development of versatile, sensor-agnostic open-source UAVs.

KW - edge processing

KW - BeagleBone AI-64

KW - YOLOv5

KW - UAV

KW - open-source

U2 - 10.1109/IWCMC61514.2024.10592445

DO - 10.1109/IWCMC61514.2024.10592445

M3 - Conference contribution

BT - Proceedings of the 20th International Wireless Communications & Mobile Computing Conference

PB - IEEE

Y2 - 27 May 2024 through 31 May 2024

ER -

Edge-accelerated UAV operations: a case study of open source solutions

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this