123. IMPROVING DRIVER SAFETY THROUGH SMARTPHONE-BASED INTELLIGENT VEHICLE APPLICATIONS

Department: Electrical & Computer Engineering
Faculty Advisor(s): Mohan Trivedi

Primary Student
Name: Derick Arnold Johnson
Email: dajohnso@ucsd.edu
Phone: 858-822-0076
Grad Year: 2012

Student Collaborators
Minh Van Ly, m6ly@ucsd.edu

Abstract
Smartphones are currently seen as one of the biggest distractions to modern day drivers. Conversely, as processing power, ubiquity, and social interconnectivity grows, so do opportunities for harnessing these sensor-packed devices for increasing driver's safety. While others may focus on decreasing visual interactions with these devices, we are focusing on sampling, analyzing and classifying driving data collected from the accelerometer, gyroscope, camera, GPS and other currently available sensors. Our Intelligent Vehicle Applications (IV-Apps) can process sensor data for a wide range of utilities including driving style recognition. Current Driver Assistance Systems (DAS) use driving style recognition, an important area of study, to provide feedback to the driver. In this study, we will attempt to improve the accuracy of the driving style recognition IV-App using new methods for segmenting and classifying data in higher dimensions.

« Back to Posters or Search Results