Vehicle Detection and Localization for Intelligent Transportation Systems


Bakhuraisa, Yaser and Abd. Aziz, Azlan and Tan, Kim Geok (2022) Vehicle Detection and Localization for Intelligent Transportation Systems. In: 2nd FET PG Engineering Colloquium Proceedings 2022, 1-15 December 2022, Multimedia University, Malaysia. (Unpublished)

[img] Text
24_Yaser Bakhuraisa_PPM Extended Abstract_Second Half 2022.pdf
Restricted to Registered users only

Download (112kB)


Background – Acurate localization is considered and essential requirement for Intelligent Transportation System (ITS), especially when the vehicle is existing in the urbane environment. However, the localization process can be established by either global or local technologies. The local localization technologies are used as an alternative to global where the performance of the global technologies is limited in urban environments. However, wireless network-based localization provides several advantages over the local localization techniques. However, there are challenges still remain to meet in wireless-based localization. The literature reported that, the geometricbased method is the most promising to meet these challenges. Purpose – The purpose of the research, is to propose a geometric-based localization algorithm to estimate the vehicle position n urban environment. To improve the performance of the proposed algoritm, we will employ a machine learning to estimate the distance between the vehicle and base station. Design/methodology/approach – In this wrok we will present analysis of the propagation paths of mm-wave in urban environment. We will use ray tracing technique to predict the characteristics of the paths. Based on this analysis, we will present two methods to etimat the distance. The first method is statistical, while the second method based on machine learning. The estimated results of the two methods will be compared. In the least phase of this work, we will ropose a geometric ased loclisation technique by utilizing the estimated distance form the previous phase. We will also apply machine learning to estimate the vehicle position directly and compare its performance with the geometric algorithm. Findings – The obtained results show that, the statistical a significant amount of error for distance stimation for both LOS and NLOS scenarios. This error is reduced by using machine learning. However, although the machine learning auotperform the staticacl method, but its performance still needs to be improved to achive the accuracy level of localization. Research limitations– The obtained data are collected need to do some preprocessing, filtering. Originality/value – present an effective ranginig model using RSS.

Item Type: Conference or Workshop Item (Other)
Uncontrolled Keywords: Intelligent transportation systems
Subjects: T Technology > TE Highway engineering. Roads and pavements > TE210-228.3 Construction details Including foundations, maintenance, equipment
Divisions: Faculty of Engineering and Technology (FET)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 16 Feb 2023 08:10
Last Modified: 16 Feb 2023 08:10


Downloads per month over past year

View ItemEdit (login required)