Live 3D Digital Twins Powering Connected Autonomy

Abstract:

While autonomous vehicles (AVs) have shown promise in normal driving conditions, handling complex scenarios is significantly challenging. This talk explores how live 3D digital twins, real-time high-fidelity 3D representations of the world, can improve both safety and efficiency for autonomous vehicles.

The talk will discuss:

•⁠  ⁠Augmenting real-time vehicle perception with live 3D views from road-side sensors to overcome line-of-sight limitations .

•⁠  ⁠Utilizing centralized motion planning to improve throughput at intersections with coordinated AV movements.

•⁠  ⁠Enabling seamless handovers between multiple edges, ensuring smooth and reliable operation as AVs navigate through multiple intersections.

Bio:

Fawad Ahmad is an assistant professor in the Computer Science department at RIT. His research focuses on building perception infrastructure to construct and update live 3D digital twins. Moreover, he is also interested in building end-to-end cyber physical systems that use these twins to enable novel and exciting capabilities. He received his PhD from the University of Southern California where he worked with Ramesh Govindan.