Modelling and Identification of Spatiotemporal Systems

The increased sensing mechanisms and data gathering in non-Engineering disciplines such as life science, medicine and the social sciences have resulted in a demand for more sophisticated quantitative analyses. Often, these analyses demand a greater understanding of the systems behavior, particularly from a dynamical systems perspective, such as causality and predictability. One class of data are of spatio-temporal in nature where measurements are made at different spatial locations and over time.

The aims of this project is to derive novel spatio-temporal modeling methods demanded by specific application case studies that have spatio-temporal data. It will build on the group’s success in developing such models for intracranial EEG signals, novel array sensor signals of ECG and EMG, and dynamics of social conflicts. The spatiotemporal model estimation will be based on a statistical inference framework and will utilise Kalman filtering concepts.