HomeEvents - Talks        <>

Invited Talk - Kai Li

ABSTRACT:
In UAVs-aided sensor networks, the flexibility of the UAV and favourable line-of-sight connection between the UAV and ground sensors are exploited to improve data reception. However, the ground nodes can suffer from buffer overflows, because of poor cruise control of the UAV and lossy airborne channels when the UAV is far away from the ground nodes. In this talk, we discuss a new deep reinforcement learning based onboard control framework, where the cruise control of the UAV and its schedule of data aggregation can be learned online. Moreover, deep deterministic policy gradient (DDPG) can be adopted onboard at the UAV to deliver effective continuous control operations and communication decisions, using its substantially outdated knowledge of the ground nodes, network states and channel conditions.


BIO:
Dr. Kai Li is a senior research scientist at CISTER Research Centre, Portugal. He is also a research fellow with Carnegie Mellon Portugal Research Program. He received the B.E. degree from Shandong University, China, the M.S. degree from The Hong Kong University of Science and Technology, and the Ph.D. degree from The University of New South Wales, Australia. Dr. Li was a postdoctoral research fellow at The SUTD-MIT International Design Centre, The Singapore University of Technology and Design, Singapore. He was a visiting research assistant at ICT Centre, CSIRO, Australia. From 2010 to 2011, he was a research assistant at Mobile Technologies Centre with The Chinese University of Hong Kong. His research interests include vehicular communications and security, resource allocation optimization, Cyber-Physical Systems, Internet of Things (IoT), human sensing systems, sensor networks and UAV networks.

,

Kai Li

Speaker