Gyroscope explorer terrain angles classification

Yalung, Christofer, Al-Majeed, Salah, Adolfo, Cid Mathew , Karam, Jalal and Mirtskhulava, Lela (2017) Gyroscope explorer terrain angles classification. In: East-West Design & Test Symposium (EWDTS), 14-17 Oct. 2016, Yerevan, Armenia.

Full content URL: https://doi.org/10.1109/EWDTS.2016.7807679

Full text not available from this repository.

Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive

Abstract

Mobile Applications (Apps) offer numerous advantages related to entertainment, communication, monitoring and sensing to name a few. In this study, a Gyroscope Explorer Apps is employed for data gathering of azimuth, pitch, and roll. The mobile phone is carried by Lego Mindstorms (EV3), in which it travels the ladder into the different angles: 4.13°, 7.77°, 10.81°, and 12.80°. The data collected was classified into eight classes: 4.13°uphill, 4.13°downhill, 7.77° uphill, 7.77° downhill, 10.81° uphill, 10.81°downhill, 12.80° uphill, 12.80° downhill. Artificial Neural Network is used to classify the angels and the orientation of the vehicle-uphill or downhill. A total of 718 data collected and divided into three sets (azimuth, pitch and roll). The number of neurons in the hidden layer is set to 10, yielding a 100% accuracy classification.

Keywords:app, artificial neural network, Mobile Robot Navigation.
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
Divisions:College of Science > School of Computer Science
ID Code:46626
Deposited On:23 Sep 2021 15:58

Repository Staff Only: item control page