Thesis - Open Access
Master of Science (MS)
Department / School
Electrical Engineering and Computer Science
Collision, GPS, Pedestrian Safety, Smartphone, Wifi, Wifi-Direct
Pedestrian distraction from smartphones is a serious social problem that caused an ever increasing number of fatalities especially as virtual reality (VR) games have gained popularity recently. In this thesis, we present the design, implementation, and a pilot study of WiPedCross, a WiFi direct-based pedestrian safety system that senses and evaluates a risk, and alerts accordingly the user to prevent traffic accidents. In order to develop a non-intrusive, accurate, and energy-efficient pedestrian safety system, a number of technical challenges are addressed: to enhance the positioning accuracy of the user for precise risk assessment, a map-matching algorithm based on a Hidden Markov Model is designed; to minimize energy consumption, an adaptive scheme is developed that dynamically activates the GPS module of a phone according to pedestrian walking speed and the locations of nearby crosswalks; to suppress false alarms, a novel algorithm is developed to accurately identify the user-phone-viewing activity so that collision probability assessment is triggered only when the pedestrian is walking while viewing his or her phone. The prototype of the proposed framework is implemented on an Android platform for a pilot study to evaluate feasibility, reliability, and validity of WiPedCross. Extensive experiments are performed in a parking lot and the results demonstrate that WiPedCross assesses the collision probability effectively and provides warning to the user in a timely manner. The system modules of the proposed framework are expected to benefit numerous other pedestrian safety apps.
Library of Congress Subject Headings
Pedestrians -- Safety measures.
Pedestrian accidents -- Prevention.
Includes bibliographical references
Number of Pages
South Dakota State University
Shrestha, Aawesh Man, "Design, Implementation and a Pilot Study of Mobile Framework for Pedestrian Safety Using Smartphone Sensors" (2018). Electronic Theses and Dissertations. 2669.