Vehicle Maneuver Detection with Accelerometer-Based Classification.

Javier Cervantes-Villanueva,Daniel Carrillo-Zapata,Antonio Skarmeta,Mercedes Valdes-Vela,Fernando Terroso-Saenz

doi:10.3390/s16101618

Javier Cervantes-Villanueva, Daniel Carrillo-Zapata + Show 3 more

Open Access

PDF Available

https://doi.org/10.3390/s16101618

Copy DOI

Export

Save

Cite

Journal: Sensors	Publication Date: Sep 29, 2016
Citations: 26	License type: CC BY 4.0

Affiliation: University of Murcia

Abstract
Highlights/Summary
Full-Text PDF
Similar Papers

Abstract

Listen

In the mobile computing era, smartphones have become instrumental tools to develop innovative mobile context-aware systems. In that sense, their usage in the vehicular domain eases the development of novel and personal transportation solutions. In this frame, the present work introduces an innovative mechanism to perceive the current kinematic state of a vehicle on the basis of the accelerometer data from a smartphone mounted in the vehicle. Unlike previous proposals, the introduced architecture targets the computational limitations of such devices to carry out the detection process following an incremental approach. For its realization, we have evaluated different classification algorithms to act as agents within the architecture. Finally, our approach has been tested with a real-world dataset collected by means of the ad hoc mobile application developed.

Highlights

IntroductionSmartphones have been the center of most of the technological advances for the last decade
Smartphones have been the center of most of the technological advances for the last decade.As a result, they are currently equipped with several embedded sensors like a GPS, accelerometer, gyroscope or magnetometer
We have considered three different data-driven and supervised classification algorithms, namely random forest (RF), support vector machines (SVM) and fuzzy rule-based classifier (FRC)

Summary

Introduction

Smartphones have been the center of most of the technological advances for the last decade As a result, they are currently equipped with several embedded sensors like a GPS, accelerometer, gyroscope or magnetometer. They are currently equipped with several embedded sensors like a GPS, accelerometer, gyroscope or magnetometer This makes them suitable enablers to capture a wide range of contextual features, like weather [1] and traffic [2] conditions, or user behaviors [3]. Smartphones are used to enrich the sensing features of traditional situation-aware driving assistance systems (SADASs) [9] This allows one to capture a wider range of driving events resulting in safer and more comfortable trips

Methods

Results

Conclusion