Title: Towards Self-Driving Vehicles: Different Approaches

Presenter: Gellért Máttyus, PhD, Senior Deep Learning Engineer – Team Leader at Continental Automotive, Budapest

Abstract: The last decade has brought enormous advancements in the field of machine learning and computer vision. One of the huge impact applications is self- and automated-driving vehicles. Deployment of these technologies will result in safer road traffic, lower transportation costs and more free time used otherwise for driving. In this presentation I will introduce different approaches with their technological challenges to implement self (automated) driving systems. First, operating Level 4 self-driving cars without a (safety) driver behind the wheels in limited geographic areas. Second, installing privately owned cars with driver assistance systems providing more and more autonomy. Last, deploying slow moving, lightweight delivery robots to the sidewalks and streets which can be teleoperated. These are particularly interesting problems for the field of computer vision, as they involve many of the major computer vision tasks.