Scene Change Detection with Radar

Automatic Scene Change Detection deals with the problem of identifying changes in the physical environment of importance to a particular application. For most applications, changes in lighting conditions, or weather changes, or seasonal changes, are not important. However, bringing or removing objects from the scene, or other changes in scene geometry, are important changes that must be detected. In case of telecom, power, or construction industries, displacement of equipment or installation, e.g., due to strong wind, is a change that must be detected.

The goal of this project is to study in-depth and evaluate the ability of a stationary radar to perform Automatic Scene Change Detection. Further, research will be performed on combining pre-built LiDAR [1] generated 3D pointcloud with radar [2] periodically scanning the scene. Major algorithmic steps include

• alignment of dense LiDAR and radar pointclouds,
• semantic understanding of the 3D scene,
• logic for identifying scene changes in radar data and
• highlighting correspondent area in the dense LiDAR based 3D pointcloud.

This thesis project is sponsored by Ericsson Research and will be jointly supervised by Örebro university and Ericsson Research in Stockholm.

Deliverables expected at the end of the thesis project include clearly documented working code, a final thesis report, and a live demonstration to an Ericsson audience.

Good programming experience with Python/C/C++ is essential for the project. Experience with working with sensor data and ROS is beneficial.

[1] https://leica-geosystems.com/products/laser-scanners/scanners

[2 ] https://www.sensrad.com/technology