IEEE ICMA 2022 Conference
Dense and Semantic SLAM via Deep Learning and Polarization Imaging
Hong Zhang, Ph.D.
Department of Electronic and Electrical Engineering
Southern University of Science and Technology (SUSTech)
This talk will focus on recent research in robot SLAM (simultaneous localization and mapping) and, specifically, on the subject of building a dense and semantic representation of a scene through visual sensing. In autonomous robot navigation, an environment representation must be sufficiently dense for the robot to perform such critical tasks as path planning and collision avoidance. A recent trend in robot SLAM calls for this representation to contain semantic information about the objects in the scene in order to facilitate map construction and manipulation and to support human-robot interaction. Upon reviewing state-of-the-art research in dense and semantic SLAM, I will present our recent work on using deep learning and polarization imaging as a way of achieving dense and semantic mapping.
Prof. Hong Zhang received his B.S. from Northeastern University (Boston) in 1982, and Ph.D. from Purdue University in 1986, both in Electrical Engineering. Subsequently he conducted research at the University of Pennsylvania as a post-doctoral fellow before he joined the Department of Computing Science at the University of Alberta, Canada where he worked for over 30 years. Since October 2020, he has been a full-time faculty member in the Department of Electronic and Electrical Engineering at the Southern University of Science and Technology, in Shenzhen, China.
Professor Zhang’s research interests include robotics, computer vision, and image processing, with over 200 publications in these areas. For the past 15 years, he has expended considerable effort in the study of mobile robot navigation. He was a principal investigator in the NSERC Canadian robotics network (NCRN) (2018-2023) - whose mandate is to develop the science and technologies to allow mobile robots to work in challenging environments, and to generate and communicate critical information to humans. He was also a PI with Centre for Autonomous Systems (2018-2022), a research initiative funded by Alberta government, to study issues and explore opportunities in autonomous and self-driving technologies. Among his many professional services, Professor Zhang is currently on a three-term as the Editor-in-Chief of IROS Conference Editorial Board (2020-2022), a flagship conference of the IEEE Robotics and Automation Society. In recognition of his research accomplishments, Professor Zhang is elected Fellow of the IEEE, and Fellow of the Canadian Academy of Engineering.