Research On Airborne Multi-sensor Information Fusion

There are many research aspects in airborne multi-sensor information fusion system. This dissertation focuses on the practical problems of engineering applications, such as system modeling, resource allocation, target recognition and target state estimation. The main contents and results in this dissertation are as follows.Based on the fuzzy set theory and Petri-Net theory, the Fuzzy-Petri-Net model for airborne data fusion system is established. Meanwhile, a Color-Petri-Net (CPN) for airborne multi-sensor information fusion system is also established.The resource allocation algorithm for multi-sensor system based on liner programming is well studied and established, aiming at the practical airborne fire control and guide system. The value of effectiveness for liner programming is the ratio of the target's relative value to the mean time of tracking, that is, the overall value of the sensors locked in each time unit is served as the objective function. The liner programming is practically feasible since the establishment of target priority and mean time for tracking.For target recognition, evidence combination equation is applied with appropriate revise. There are two kinds of combination with respect to the target's attributes which are correlative or not. In the case of correlative attributes, weights are used for evidence combination, which could be derived from information quantity or independence weight. The multi-attribution decision making model has been used in calculating those weights, with the analysis method for its sensitivity.The structure and algorithm for airborne multi-sensor tracking system are also investigated. A modified probability data association (PDA) algorithm has been adopted for estimating the clutter density based on the transcendent information of each history scan for obtaining the more precise parameters of clutter and target states. The central state estimation algorithms for twin-sensor synchronized measuring and asynchrony measuring are put forward respectively. The PDA filter, which is quite sensitive to the density of spurious responds, however, with the proposed fusion-tracking gate, is well solved. Simulation results verified those methods. For bettering the tracking quality of multi-target tracking system, the maximum likelihood estimation (MLE) is used for point data fusion of airborne radars and infrared sensors. With flexible tracking gate and unblocking method,...