INCORPORATING TRAJECTORY INFORMATION IN RANDOM MATRIX ELLIPTICAL EXTENDED TARGET TRACKING

Date

2024-9-02

Author

Şahin, Kurtuluş Kerem

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

38
views

0
downloads

Cite This

This thesis focuses on Extended Target Tracking (ETT) using Random Matrix Methods (RMM), which provide enhanced estimations of target size and movement in tracking systems. Traditional methods often miss crucial trajectory details, which, if considered, could improve tracking performance. To address this issue, we have developed two new RMM-based models. The first, the trajectory-aligned model is designed for targets moving in a consistent direction, ensuring that the orientation aligns with the trajectory. The second, the drifting model is for targets whose orientation deviates from their heading direction. Utilizing the variational Bayes (VB) method, we obtain posterior densities by performing analytical and iterative steps for both models. This methodological choice ensures that our models not only deliver precise tracking results but also operate efficiently in real-time applications. Extensive testing on both simulated and real-world data has proven that our models effectively outperform current methods in handling drifting and trajectory-aligned targets. These tests confirm the flexibility and efficiency of our models under diverse conditions. The demonstrated success of our models in both simulated and real environments underscores their potential to significantly enhance current standards in extended target tracking.

Subject Keywords

Extended Target Tracking, Random Matrix, Maneuvering Motion Models, Variational Bayes, Automotive Radars

URI

https://hdl.handle.net/11511/111382

Collections

Graduate School of Natural and Applied Sciences, Thesis