Association of the Tracked Trajectories and Marks by the Attraction Method During Secondary Processing of Radar Information
The article proposes a new algorithm for associating the tracked trajectories and newly received marks by coordinates during the secondary processing (track-while-scan) of radar information. It is known that the biggest difficulties arise when associating in dense groups, that is, when the distance...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | Russian |
| Published: |
Educational institution «Belarusian State University of Informatics and Radioelectronics»
2022-12-01
|
| Series: | Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki |
| Subjects: | |
| Online Access: | https://doklady.bsuir.by/jour/article/view/3503 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The article proposes a new algorithm for associating the tracked trajectories and newly received marks by coordinates during the secondary processing (track-while-scan) of radar information. It is known that the biggest difficulties arise when associating in dense groups, that is, when the distance between the trajectories is commensurate with the errors in measuring their coordinates. Usually, well-known methods for solving the assignment task are used, for example, the Hungarian algorithm and similar. A common disadvantage of these methods is a rapid increase (in proportion to the third or fourth degree of the number of trajectories) in the time for solving the task. The article proposes to use the “attraction” method to increase the speed of calculations. The proposed algorithm simulates the “attraction” of all trajectories to all marks and the mutual “repulsion” of all trajectories from each other (the position of the trajectories is extrapolated to the time of location of the marks.) The conditional “movement” of the trajectories is simulated step by step until a set approach to any marks happens. Comparative modeling of the attraction algorithm and the Hungarian algorithm in the case of equal number of trajectories and marks showed that the qualitative characteristics of the algorithms are approximately the same, but the execution time for the attraction algorithm grows more slowly than for the Hungarian algorithm (in proportion to the square of the number of trajectories). Therefore, with a large number of them (more than 100–300), the attraction algorithm is executed much faster. Obviously, with the corresponding adjustment of the value and dimensions of the parameters, the new algorithm can be used to solve other assignment tasks. |
|---|---|
| ISSN: | 1729-7648 |