Dissimilar Sensors Fusion of Airborne Early-warning and Surveillance System for Multi-targets Joint Detection, Tracking and Recognition

To realize the multi-targets detection, tracking and identification by the dissimilar sensors fusion in the air early-warning and surveillance system, the multi-targets joint detection, tracking and recognition strategy based on the random finite set model for the dissimilar sensors fusion is studied. Through the unified description of the target’s kinematics state and target recognition attribute state, the multi-targets state is modeled as a global state described by a random finite set. By analyzing the kinematic sensor and attribute sensor models, the various sensors are modeled as a global sensor, and then each dissimilar sensor measurement is modeled as a global measurement described by a random finite set. According to the global state and the global measurement model, the process of multi-targets joint detection, tracking and recognition by dissimilar sensors fusion is described as a Bayesian filtering process, and the corresponding strategy of multi-targets joint detection, tracking and recognition has been proposed. Simulated experiment is given to prove the correctness and effectiveness of the proposed strategy.