Application of data and information fusion

Abstract

Data and information fusion is emerging as an important field of multi-disciplinary study. This is engendered by increased data and information flow, as well as improvement in communication, computing and sensor technology. In this project, we first survey high level information fusion, a research topic that is comparatively less mature than low level information fusion. Relevant application areas are also discussed. Secondly, we develop an approach for intent inference, where knowledge on preceding activities of a target of interest is utilized to predict its future action. Simulation test results demonstrate that our approach provides timely inference, which is useful for human cognition and decision making. Thirdly, we derive an algorithm for manoeuvring target tracking via the combination of existing techniques. Experimental results show that superior performance can be attained at modest computational costs. Lastly, real data from application problems are used to test the feasibility of the aforementioned proposed methods.