An Intelligent Technique to Multi-sensor Data Fusion in Target Tracking

Combining the results of multiple sensors can provide more accurate information than using single sensor. In this paper, we develop fuzzy clustering approach to data association and track fusion in multisensor multi-target environment. The proposed approach uses the fuzzy clustering means algorithm to get the degree of membership of new tracks to existing tracks. Unlike existing approaches, in which the membership functions are fixed a priori (determined empirically), we generate optimal membership functions from the data using the fuzzy clustering means algorithm. More specifically, the values of the membership functions change according to the relative positions of the targets with respect to the sensors; this adaptation to the current state of the environment leads to far better/accurate results. Furthermore, our proposal can handle different types of information without excessive computation; indeed, it reduces considerably the computational complexity compared to existing schemes.