Training of a deep learning algorithm for quadcopter gesture recognition

College

Gokongwei College of Engineering

Department/Unit

Mechanical Engineering

Document Type

Article

Source Title

International Journal of Advanced Trends in Computer Science and Engineering

Volume

9

Issue

1

First Page

211

Last Page

216

Publication Date

1-1-2020

Abstract

Traditional methods to control Unmanned Aerial Vehicles are unintuitive and susceptible to radio interference. Recent research has shown that hand gestures are the most intuitive method for quadcopter control. Also, deep learning in the form of a convolutional neural network is a more compatible approach to gesture recognition than other methods. This paper presents the design, and training of a deep learning convolutional neural network for gesture recognition and tracking of a quadrotor Unmanned Aerial Vehicle. The neural network was coded in Python using the Keras library and was trained on a laptop computer. Inference was performed on a Raspberry Pi 4 computer that is intended for use as a companion computer aboard a quadcopter. © 2020, World Academy of Research in Science and Engineering. All rights reserved.

html

Digitial Object Identifier (DOI)

10.30534/ijatcse/2020/32912020

Disciplines

Mechanical Engineering

Keywords

Drone aircraft—Control systems; Neural networks (Computer science)

Upload File

wf_no

This document is currently not available here.

Share

COinS