Multi-Target Tracking For General Aviation Based On Random Finite Sets

At the present,the opening of low-altitude airspace has drawn a considerable attention in China,and the general aviation will play a key role in the national economic constructions and social developments.Reliable surveillance of the low-altitude targets remains a crucial problem.To solve the technical problem of surveillance with the opening low-altitude airspace of China,multi-target tracking approaches are studied in this dissertation.The researches in this dissertation are of significance for the flight service and the development of general aviation.In this dissertation,the approaches of tracking multiple targets for general aviation in low-altitude are based on multi-sensor surveillance network.Three problems are studied and can be summarized as follows:1.The existing GM-PHD filter can not give satisfactory results in the low-altitude airspace with dense clutter.The measurements optimal assigned GM-PHD filter is proposed for multi-target tracking,which distinguishes the measurements by using a maximum likelihood adaptive gate.In the new approach,survival targets and birth targets update the PHD estimation using the optimal assigned measurements respectively.The simulation results show that the proposed approach works effectively although in the dense clutter environment.For reducing the computational cost of the GM-CPHD filter,a fast algorithm is proposed,which reduces the clutter measurements using the max likelihood adaptive gate.The simulation results show that the proposed approach reduced the computational cost obviously,while obtained a similar performance.2.More measurements are generated by the target per observation interval,when the target is detected by a high resolution sensor,or there are more measurement sources on the target surface.How to accurately and effectively partition the measurements of multiple extended targets is a crucial problem.In this dissertation,the affinity propagation clustering is introduced into measurement partitioning for extended target tracking,and the adaptive gate is used to remove the clutter measurements,which makes the affinity propagation clustering being capable of partitioning the measurement in dense clutter environment with high accuracy.The simulation results are presented to demonstrate the performance of the proposed algorithm,which obtained an improved performance,while reduced the computational complexity obviously.3.The approaches of tracking multiple targets for general aviation are based on multi-sensor surveillance network.As describing the systematic bias of radar by random finite sets,solutions for multi-radar bias estimation without priori association are proposed both in geometry projection coordinates and earth-centered earth-fixed coordinates.And a cooperative estimation of radar systematic bias is proposed by using ADS-B surveillance data as the high-accuracy reference source.The simulation results are provided to verify the effectiveness and improved performance of the proposed approach for systematic bias estimation.