Radar Reticence is a start-up that develops next-generation radar technology for safer and more autonomous vehicles. Built on deep Swedish radar expertise and research from Halmstad University, our patented software-based solution makes automotive radar more robust against interference – one of the fastest-growing challenges in advanced driver assistance and autonomous mobility.

We are now entering an exciting growth phase: moving from advanced research and validated testing toward chip integration, customer proof-of-concepts and commercial deployment with leading actors in the mobility and radar ecosystem. In 2026, Radar Reticence secured long-term financing through a new investment round and was awarded EIC Accelerator funding from the European Union.

This thesis work will be in our office in Linköping.

The Thesis

In our data collection campaigns we have synchronized data from both the new radar and commercial stereo cameras. One of the ideas with the setup is to use object detection in the stereo cameras to annotate the radar signal for usage in a Machine Learning framework. The first step here is to refine the built-in stereo algorithms and also to combine data from all the four cameras mounted on the car. That is to combine and fuse results from 12cm baseline in each camera with the larger baseline of 50-75cm between the stereo cameras.

The algorithm development will be in C/C++ based on collected data and could be done both as an individual project or as joint project with two students.

Qualifications

Master of Science student with knowledge in C/C++ and machine learning as well as general image processing.