Autonomous Scalable Knowledge Extraction and Decision Making for Complex Systems and Dynamic Environments

About the Project

This PhD research is focused on multi-sensor data fusion for decision making. The project aims at dealing with large volumes of data and high dimensional systems. We have two case studies in particular: intelligent transportation systems and search and rescue. In both case studies there are elements of data collection via intelligent sensing, followed by detection of the important events. In the light of autonomous decision making, the project aims at developing machine learning algorithms for knowledge extraction from data, especially LIDAR, radar and others. Tracking is also part of this task since it require monitoring the areas of interest and autonomous decision making.

Possible applications to this research are: Unmanned Aerial Vehicles (UAVs) sent to perform a mission, e.g. search and rescue operations, intelligent transportation systems and wireless sensor networks. The data can vary from image and video, GPS, GSM to radar and other types. The challenges both from modelling, sensing, filtering and decision making point of view can be considered.

Potential methods involve a broad range of machine learning methods, including sequential Monte Carlo methods, Gaussian processes and Bayesian compressed sensing.

Applicants from different backgrounds are encouraged to apply depending on the specific nature of the project which can also be adapted to students’ interests and experience.

Prospective candidates should have an MSc, MMath or MEng in mathematics, statistics, physics, aerospace engineering, signal processing, electrical engineering or a related subject.

Funding Notes

Applicants can apply for a Scholarship from the University of Sheffield but should note that competition for these Scholarships is highly competitive. it will be possible to make Scholarship applications from the Autumn with a strict deadline in late January 2020. Specific information is available at: View Website