Job Description

We live in a world of interconnected networks of autonomous sensing systems, catering to diverse applications in fields of automotive, satellites, robotics and many more. These distributed systems will have to fuse a combination of batch, streaming and intermittently-available datasets (typically from multi-modal sensors), and make prudent inferences using minimal communication with neighboring nodes. In view of energy-efficient solutions (i.e., minimizing both communication, processing and memory resource), these mobile systems will be autonomous and will rely heavily on on-board inference. In this project, the PhD candidate will focus on developing distributed multi-modal sensor fusion algorithms for hybrid datasets (batch, streaming and intermittently-available), with applications to distributed autonomous systems. In particular, statistically robust, probabilistic machine learning algorithms will be explored for time-varying graphs, which naturally arises in mobile autonomous systems, tackling challenges of PNT, distributed inference, control and learning.

Follow @eurasip