Differentiating transportation modes using Bluetooth sensor data Public Deposited

http://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/8w32rb07f

Descriptions

Attribute NameValues
Creator
Abstract or Summary
  • State departments of transportation, transportation consultants, and other transportation agencies are always in need of data that can be used to better understand how different modes of transportation use the road and highway systems. A variety of automatic data collection technologies have been used to gather these data including video cameras, inductive loop detectors, license plate recognition, as well as wireless-based technologies such as infrared. These technologies have different capabilities with respect to the amount of information that can be derived from the collected data. Regardless of the richness of the collected data, the majority of the available technologies focus on collecting vehicle-based data because they either do not have the capability to collect data from other travel modes (e.g., bicycles and pedestrians), or may need to be deployed differently to support this capability (e.g., video technology). One type of wireless-based data collection system that has been deployed recently is based on Bluetooth technology. A key feature of Bluetooth-based data collection systems that makes travel mode identification feasible is that the Bluetooth-enabled devices within vehicles are also present on bicyclists and pedestrians. The main objective of this research was to explore the feasibility of utilizing the information contained in data collected by Bluetooth-based data collection units (DCU) to automatically identify three different modes of transportation (i.e., motor vehicles, bicyclists, and pedestrians) travelling through an intersection. To accomplish this objective, a methodology was developed that included three controlled data collection experiments and one uncontrolled data collection experiment where data were gathered from Bluetooth-enabled devices using several Bluetooth DCUs. The main performance metric utilized was the duration of travel which was calculated from the time-stamped MAC address data collected by the Bluetooth DCUs. The clustering methods k-Means, Fuzzy c-Means, and Partitioning Around Medoids were applied to the overall duration of travel data to distinguish vehicles, bicycles and pedestrians. The results obtained in this research prove that the Bluetooth-based data collection system can be a viable approach for distinguishing different modes of transportation travelling through intersections controlled by either a stop sign or traffic light.
License
Resource Type
Date Available
Date Copyright
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Non-Academic Affiliation
Keyword
Subject
Rights Statement
Peer Reviewed
Language
Replaces

Relationships

In Administrative Set:
Last modified: 10/27/2017

Downloadable Content

Download PDF
Citations:

EndNote | Zotero | Mendeley

Items