Assistant Professor at University of Illinois Urbana-Champaign
Haitham’s research focuses on building internet-of-things (IoT) systems and technologies that deliver new capabilities and applications that were never possible before. His inventions range from new sensors that enable self-driving cars to see through fog and finding surprising ways to hack (and secure) smart home assistants like Google Home and Alexa with inaudible sound. He has also developed the world’s fastest algorithm for computing the Fourier Transform, making a major leap in the 50-year old algorithms which are used across almost all computing applications, ranging from medical imaging to GPS. This work formed the core of his PhD thesis, earning him the ACM Dissertation Award (and the Sprowls award for the best doctoral dissertation in computer science at MIT), and it was selected by Technology Review as one of the world’s top 10 breakthrough technologies in 2012.
His newest invention: a sensor that enables self-driving cars to see through fog and work in adverse weather conditions. Specifically, today’s self-driving cars rely on vision sensors (like cameras and LIDARS) and thus can only see and navigate in good visibility. However, if the visibility is low (e.g., in fog or bad weather conditions), today’s cars cannot see or navigate. Indeed, this has led to multiple incidents of self-driving cars crashing in bad weather conditions. To overcome this problem, Haitham led a team of researchers that came up with a completely different solution. Their solution relies on millimeter-wave radars. Unlike visible light, millimeter-wave radars can traverse fog and rain, and reflect off other objects in the environment before coming back to the car. Prof. Hassanieh’s sensor captures these reflections and uses them to image obstacles (e.g., other cars) in order to avoid accidents. To do this, they design a first-of-its-kind AI (Artificial Intelligence) that can image and recognize cars to avoid occlusions. Their invention has been published in the world’s most selective venue for computer vision research, and the following video shows the applications of this system.