My bachelor thesis was titled Lane-level Map Matching for Model-scale Vehicles, where I did it at the CPM Lab at RWTH Aachen, Germany.

The exposé for my topic is here:

This browser does not support PDFs. Please download the PDF to view it: Download PDF.

</embed>

This opportunity was provided by Dr.-Ing. Bassam Alrifae and my work was supervised by Simon Schäfer M.Sc., and the work is done under the Cyber-Physical Mobility Group

The map matching problem arises in the field of autonomous vehicles where the position of a vehicle is obtained using GNSS sensors on the vehicle. Like many types of sensors, GNSS data involves errors arising from the idea of how satellites work to find the position of a moving vehicle. Map matching a moving vehicle to an HD map using this distorted GNSS data is well known challenge. Many algorithms where implemented and tested to address this specific problem.

An approach using Hidden Markov Model algorithm was implemented and tested by Newson and Krumm, 2009 which was successful and paved the road for many papers later that addressed this problem.

The work be Newson and Krumm is considered to be road-level map matching.

In this thesis, the objective is to implement the HMM algorithm so support lane-level map matching using GNSS data and data acquired by the vehicle itself such the heading, yaw rate etc.