Sensor-based multi-target trajectory fusion method in target tracking

Abstract

The invention provides a sensor-based multi-target trajectory fusion method in target tracking, and relates to the technical field of target tracking. The method comprises: in a target tracking process, firstly, judging whether detection data of different sensors come from the same target; adopting an adaptive trajectory association algorithm; selecting a corresponding association algorithm from aspeed-based double-threshold track association algorithm and a nearest domain track association algorithm by judging the track state; and then track fusion is carried out on tracks from the same target, and a weighted track fusion algorithm based on local state estimation is adopted: the local state estimation of each sensor is utilized to complete the weighted combined filtering of multiple sensors, so that the error problem caused by the precision change of the sensors is solved, and the track fusion precision is improved. According to the method, the target tracks can be effectively fusedin a dense or sparse environment of the target tracks, and the target track detection precision of the sensor is improved.

Description

technical field [0001] The invention belongs to the technical field of sensor target tracking, in particular to a sensor-based multi-target trajectory fusion method in target tracking. Background technique [0002] The target tracking system is to extract the trajectory after preprocessing the sensor detection data, and determine the ownership relationship of each target trajectory through the trajectory correlation algorithm, so as to achieve the degree of trajectory fusion of the same target trajectory, and to this extent obtain accurate target trajectory, which lays the foundation for trajectory prediction. The trajectory correlation technology solves the uncertainty of the trajectory origin, and the trajectory fusion technology can improve the accuracy of target tracking and solve the problem of target state uncertainty. Therefore, trajectory association and trajectory fusion techniques are very important in target tracking technology. [0003] Target tracking technolo...

AI Technical Summary

Problems solved by technology

When the trajectories intersect, the double-threshold trajectory association algorithm (Double-Threshold Trajectory Association, DTTA) uses the quadratic threshold method. Compared with the NNTA algorithm, it has a higher association rate and a lower number of false-miss associations, but The complexity of the model of the algorithm will increase the amount of calculation

Images