In this paper, a data association problem is investigated in which a multi-platform detecting multi-target scenario is considered. A PointNet-based data association algorithm for matching measurements and targets is proposed which can deal with the measurement noises, false alarms, missed detections due to environmental interference and limitations on sensor resolution, as well as unknown target motions. The constantly changing number of targets and measurements are also taken into account. First, the data preprocessing method is provided. Then, a PointNet-based data association algorithm is designed by which the data association probabilities between the real targets and the measurements are given. Then, the effectiveness and the accuracy of the proposed algorithm are proved by the test experiments.
A PointNet-Based Multi-platform Multi-target Data Association Algorithm
Lect. Notes Electrical Eng.
30.10.2021
10 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Multi-target data association algorithm based on tracking-differentiator
British Library Online Contents | 2012
|FGAs-based Data Association Algorithm for Multi-sensor Multi-target Tracking(P)
British Library Online Contents | 2003
|Personalized HRIR Based on PointNet Network Using Anthropometric Parameters
British Library Conference Proceedings | 2023
|