On the Efficiency of Running Machine Learning Tasks for Drone-Based Target Tracking : Cloud-Based vs. Drone-Based 


Vol. 43,  No. 1, pp. 143-151, Jan.  2018
10.7840/kics.2018.43.1.143


PDF
  Abstract

Unmanned Aerial Vehicles, otherwise known as drones, necessitate the execution of two artificial intelligence tasks while tracking arbitrary objects: (i) the detection and tracking of the target visual and (ii) the corresponding control of the drone movement. These tasks are known to require substantial computing power; however, due to the physical limitations of small-sized drones in their accessibility to computing hardware, it is often practically impossible to execute the said tasks in a satisfactory manner. Cloud offloading could alleviate such a problem, but it incurs a cost in the form of network delay. This paper thus presents a comparative analysis between the resource-poor drone setting and the resource-rich, network-delayed cloud setting in their accomplishment of an object tracking mission. Through this analysis, we quantify the drone-based system"s tracking performance achievable with existing technology, and also derive the anticipated requirements for computing resources and communication networks for a better accommodation of AI-related services based on drones.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

K. Son, D. Hostallero, D. Kim, J. Bang, W. J. Kang, S. Yoon, Y. Koo, H. Yeo, J. Ha, D. Han, Y. Yi, "On the Efficiency of Running Machine Learning Tasks for Drone-Based Target Tracking : Cloud-Based vs. Drone-Based," The Journal of Korean Institute of Communications and Information Sciences, vol. 43, no. 1, pp. 143-151, 2018. DOI: 10.7840/kics.2018.43.1.143.

[ACM Style]

Kyunghwan Son, David Hostallero, Daewoo Kim, Jihwan Bang, Wan Ju Kang, Se-eun Yoon, Yoon-pyo Koo, Hyun-ho Yeo, Jae-hyung Ha, Dongsu Han, and Yung Yi. 2018. On the Efficiency of Running Machine Learning Tasks for Drone-Based Target Tracking : Cloud-Based vs. Drone-Based. The Journal of Korean Institute of Communications and Information Sciences, 43, 1, (2018), 143-151. DOI: 10.7840/kics.2018.43.1.143.

[KICS Style]

Kyunghwan Son, David Hostallero, Daewoo Kim, Jihwan Bang, Wan Ju Kang, Se-eun Yoon, Yoon-pyo Koo, Hyun-ho Yeo, Jae-hyung Ha, Dongsu Han, Yung Yi, "On the Efficiency of Running Machine Learning Tasks for Drone-Based Target Tracking : Cloud-Based vs. Drone-Based," The Journal of Korean Institute of Communications and Information Sciences, vol. 43, no. 1, pp. 143-151, 1. 2018. (https://doi.org/10.7840/kics.2018.43.1.143)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880