SYNOPSIS
Lane detection is one of the fundamental components in advanced driver assistance systems (ADAS) in semi-autonomous vehicles in the form of lane keep assist or lane departure warning. It is also one of the key enablers for autonomous driving for it can be used as part of the lateral and longitudinal control system or for localization. Lane detectors determine lane boundaries by extracting the edges and color of paintings on the road. However, the paintings on the road may not always be visible for various reasons such as shadows, aging, road texture changes and other environmental factors which makes vision-based lane detection a challenging problem. In this seminar, we present our recent results on reliable lane detection to handle some of those challenges. In particular, we present one of our developed lane detectors which relies on conventional image processing techniques to enhance and extract features from the image stream which utilizes road geometry parameters like lane width and lane marking thickness. We also discuss a light weight deep neural network approach for feature extraction without the need to manually tune thresholds and other parameters as the network learns them indirectly from training data.
ABOUT THE SPEAKER
Tesfamichael Getahun is a 4th year PhD student and graduate research assistant at Autonomous Cooperative Control of Emergent Systems of Systems (ACCESS) Laboratory at North Carolina A&T State University. He received M.S. degrees in Robotics and Automation from Aalto University-Finland, M.Tech. degree in Electrical Engineering from Indian Institute of Technology Bombay-India and B.Sc. degree in Electrical Engineering from Bahir Dar University-Ethiopia. His research interests include visual perception, sensor fusion, planning and control for autonomous vehicles.