Chathura Jayath Gunasekara, Chamath Keppetiyagama, Nihal Kodikara, Chethaka Uduwarage, Damitha Sandaruwan, K.R. Senadheera, J.U. Gunaseela
This paper presents a ongoing research on generic sensor fusion architecture and its application to a maritime surveillance system. The importance of information fusion for various sensor types and specialized sensor fusion systems in various domains are discussed. In sensor fusion, centralized versus decentralized refers to where the fusion of the data occurs. In centralized fusion, the clients simply forward all of the data to a central location, and some entity at the central location is responsible for correlating and fusing the data. In this research we intend to correlate multiple maritime sensors such as RADAR, AIS, and other electronic object detection systems. Sri Lanka Navy has myriad of surveillance information sources such as AIS (Automatic Identification System), RADAR (Radio Aid Detection and Ranging), SONAR (Sound Navigation and Ranging), MSTT, HFSWR, AVL (Automatic Vessel Locating), HF Communication, Intelligence Data ect. So it is obvious that an information fusion methodology is needed to harness the effectiveness of multiple sensor information. An object identification pipeline is conceptualized such that an unknown object in the maritime domain is detected reducing the uncertainty of obtained information. The radical new virtual reality application has been developed to visualize the information fused from sensors and discussed in depth in addition to the introduction of the fusion process. The technologies used in developing the virtual world and incorporating the real time information in to the virtual world is presented in simplified modal for the purpose of clarity. Also design aspects and some experimental analysis developed in an applied project at the Sri Lanka Navy intended to demonstrate fusion technologies in this environment through an operative prototype are presented.