goseind / kalman-filter-dbscan

Kalman filter implementation with with DBSCAN clustering on 3d radar sensor

Kalman filter implementation with with DBSCAN clustering on 3d radar sensor

This repository demonstrates a sophisticated implementation of object tracking using a 3D radar sensor. By integrating a Kalman filter and DBSCAN clustering, this code provides a comprehensive solution for accurate and robust detection and tracking of objects in three-dimensional space.

Python algorithm implementations

• Kalman Filter 1D
• DBSCAN 3D

Jupyter Notebook documentation

docs are in german

• Interactive Jupyter Notebook
• Interactive Presentation

kalman-filter-dbscan's Issues

Implement DBScan

Implementierung des DBSCAN Algorithmus.

Smoothing

Smoothing des Kalman Filter Outputs implementieren.

Q Noise Function

Q Noise Function aus Kalman Notebook verstehen und eigene Implementation erstellen.

Make checkbox dependent on number of targets

Only print false dets when one target is selected from the list.

Motion control

Motion control beim Kalman Filter einbinden.

Adding 3D Filter source code from moodle

Cleanup GitHub Repo

Wir sollten uns demnächst auf einen Stand einigen, diesen in die main Branch übertragen und die anderen Branches ggf. löschen, damit wir später keine merge Konflikte bekommen.

Google Colab Jupyter Notebook

Erstellen eines Jupyther Notebooks mit Google Colab, um das Projekt dort zu dokumentieren.