1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Title:	A Real Time Automated Vehicle Incident Detection Using High Accuracy Global Positioning System
Authors:	Ishak Mohamad (P42465)
Supervisor:	Mohd. Alauddin Mohd., Professor Dr. Ali
Keywords:	Automated Vehicle Incident Detection Vehicle Incident Detection High Accuracy Global Positioning System Traffic accidents
Issue Date:	31-Jul-2013
Description:	The Malaysian Institute of Road Safety Research (MIROS) reported that in 2009, there were 397,330 road accidents in Malaysia and half of them were occurred on municipal roads. In order to avoid major traffic congestion and other consequent incidents, an efficient incident detection system is therefore essential. Although various incident detection systems have been deployed such as laser, multi-sensors, closed-loop detector and Global Positioning System (GPS), the systems are still not fully automated, require longer delay for detection, have high operational cost and high detection error. Moreover, GPS is only used for navigation or positioning and not fully utilized for incident detection. The objective of this research is to develop a new and efficient real time automated incident detection (AID) system with short time to detect, no false alarm and higher detection rate. The research involved three phases; data collection, AID algorithm development and verifications through simulation and drive testing. During static data collection, GPS signal availability and reliability at four locations in the study area were checked. In the mobile data collection, the GPS probe vehicle gathered various information such as vehicle position, direction, and speed while driving along municipal trunk roads in Bandar Baru Bangi, Selangor. The data were transmitted via ultra high frequency (UHF) radio band to a monitoring station (MS). The AID algorithm detected abnormal measures based on user input threshold of the vehicle's maximum speed, maximum acceleration and maximum deceleration from mobile data collection. The algorithm was also verified through simulation and the result could be viewed from the AID Interface (AIDEG) using random data for the normal and abnormal condition extracted from recorded mobile data. The system performance indicator includes time to detect (TTD), false alarm rate (FAR) and detection rate (DR). Finally, real time data from five GPS equipped vehicles along the pilot route were sent to the MS to test the performance of the AID system. The static test shows GPS availability and reliability rate of 99.9% and up to 99.7% respectively, thus the study area is appropriate for AID application. Algorithm parameters obtained from mobile data are 33.33 m/s for the maximum speed, 3.24 m/s2 for the maximum acceleration and 4.63 m/s2 for the maximum deceleration. The result for real time monitoring shows excellent value of performance indicators; 3.3 μs to 3000099 μs TTD, zero FAR and 81% and above DR. However, the performance of the system depends on the availability of GPS signal and data transmission from the vehicles to the MS.,PhD
Pages:	153
Call Number:	HE5614 .I834 2013 3
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.