As the cutting-edge technology, autonomous driving has received tremendous attention in academy and industry. It involves several inter-disciplinary fields, like artificial intelligence, machine learning, automatic control, and large scale integrated circuit. In this talk, I will focus on the AI challenges in autonomous driving, as well as our solutions. The talk includes, 1. system integration on software and hardware; 2. object detection and recognition based on deep manifold network (DMN); 3. High-precision localization based on structural SLAM in complex scenario; 4. AI chip design based on Massive, Parallel, and Volumetric (MPV) architecture.
Xiaofei He is the founder and CEO of FABU.ai, and a professor in the College of Computer Science at Zhejiang University, China. He was the former senior vice president of DIDI, and the founder of DIDI research labs and the self-driving team. He received the PhD degree from the University of Chicago, thereafter worked as a research scientist at Yahoo! Research Labs, and then joined Zhejiang University as a full professor. He has published more than 200 research papers in leading conferences/journals in the field of artificial intelligence and machine learning. His entire publications have been cited over 24,000 times. He received the National Science Funds for Distinguished Young Scholars in 2011, the most prestigious academic award in China. He is on the advisory board of the Intelligent Highway Project of Zhejiang province. He is an area chair of NIPS 2018, the leading international conference in artificial intelligence. He is a Fellow of the International Association of Pattern Recognition (IAPR) and a senior member of the IEEE.