A multi-sensor fusion tracking system based on fuzzy tracker

Multi-sensor data fusion systems have been widely used for tracking maneuverable targets, since utilizing data fusion for combing information from different sources can reduce the disadvantages of the single-sensor systems. Radar and infrared imaging are two kinds of sensors for targets detection and tracking, while radar usually works larger distance than infrared sensor and in most situations infrared imaging sensor supplies more accurate information of target than radar, for radar can be more easily affected by noise though it has more power. In addition, with infrared imaging information target maneuvere can be detected in time. In this paper, a multi-sensor data fusion system based on infrared, imaging and radar system for tracking maneuverable targets is presented. Consider the asynchronous problem of the two kinds of sensor, we use a least square technique to produce one fictitious information that is time-aligned with the next radar measurement. Given the fictitious information, we use a weighted average method to fuse this information with the radar data. Finally, we use a fuzzy logic (alpha-beta) tracker to determinate the maneuverable targets position, which does not need a maneuver detector even when tracking super maneuvere targets, and it is easy to implement. Numerical simulation results are given, which shows that the proposed multi-sensor fusion tracking system has good tracking performance.