Multi-Modal Perception from a Perspective of Hardware

Date: 2022/11/25 - 2022/11/25

Academic Seminar: Multi-Modal Perception from a Perspective of Hardware

Speaker: Xuyang Lu, Assistant Professor, University of Michigan-Shanghai Jiao Tong University Joint Institute

Time: 10:00 - 11:30, November 25, 2022 (Beijing Time)

Abstract

Autonomous driving is one of the most thrilling topics nowadays. One of the challenges for today‘s autonomous driving is the cost and complexity of the implementation of different sensors including optical, sound, radar, sonar, cameras, and more. The development of the semiconductor circuits enables the integration of sensors as well as advanced processing circuits, yet further improvement is still necessary to allow L4 or L5 level intelligence. In this talk, we will focus on two important questions related to autonomous driving hardware. First, we will dive down into the popular debate of radar and Lidar, and look into the underlining physics to see which one is going to dominate in the future. Secondly, we will discuss the trend of processing the signal in the analog domain in sensor fusion.

Biography

Xuyang Lu received his B.S. degree in electrical engineering from Rice University, Houston, TX, USA, in 2014, and the M.A. and Ph.D. degrees in electrical engineering from Princeton University, Princeton, NJ, USA, in 2016 and 2020, respectively. He joined as a Faculty Member with the University of Michigan-Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai, China, in 2021, with a joint appointment with the Department of Electronic Engineering. He is also a member of the MoE Key Laboratory of High-Speed Electronic Systems, at Shanghai Jiao Tong University. His research interests include high-speed programmable RF and mmWave integrated systems, integrated terahertz systems, integrated photonics, on-chip antenna optimization, and machine learning in analog circuit design.