Infrastructure Systems: Data Analytics for Vehicle-based Mobile Sensing Data

Scholar: Jordan Miller

Home Institution: Austin Peay State University

PROJECT

During the summer of 2018, Jordan Miller worked with Dr. Jinwoo Jang to work on classifying road lane markings that need to be repainted. Self-driving cars rely on road lane markings, and if no lanes are present then cars cannot drive themselves. This would result in more accidents on the road due to cars being unable to operate without their respective lanes. Jordan worked to find a solution to this problem by using an opensource lane detection algorithm adapted to work in ‘almost real-time’. The method includes placing a webcam on the dash of a car and having it save video in one-minute intervals. These videos are then read into the lane detection algorithm where the algorithm detects if the lanes are there or not by drawing red lines over the video. The next phase of this research includes being able to send the GPS coordinate of the lane that needs to be repainted to the local government.

Click here for the presentation. Click here to watch the video presentation.