This is a 6 months Master thesis role.
One important component of the Automated Driving vehicle navigation is the localization.
Autonomous vehicle navigation is a complex task that involves a precise perception of the environment and path planning considering known obstacles and uncertainty.
The vehicle needs to create a map of the environment and be able to operate in unstructured environments with reasonable computational cost.
Most existing localization approaches for AD are based on Differential GPS.
The task of this master work to achieve highly accurate localization without DGPS.
This should be achieved by fusion all available sensor data: vehicle odometry, GPS, IMU, Lidar, Radar and optical feature tracking in combination with information being obtained from the localization layer of the navigation map.
Below are the steps we expect the student will pass during the work on the topic:
• Literature review
• Algorithmic development of grid based sensor fusion approach
• Evaluation of the algorithm on real data
• Performance analysis of the implemented algorithm
Completed Bachelor or equivalent degree in Information Technology or Electrical Engineering
Interest in Automated Driving Basic knowledge in image processing, CV, machine learning
Programming languages: MatlabTM, C/C++
Good English skills
German language skills are not mandatory but advantageousInside this Business Group
Automated Driving Group : Zero accidents. Mobility for all. Intel is collaborating with the world's leaders in automotive design and technology to turn visionary concepts for automated driving into reality. The Automated Driving Group (ADG) is architecting transportation for a better life and a safer world. We accelerate innovation and adoption of smart, connected, transformative, market leading automated driving solutions by delivering high performance SOC’s, modules, software and reference designs.