The working title is "an imaging- and odometry-based localisation and guidance system for land autonomous systems". The abstract, translated:

Autonomous land-traversing robots are faced with the problem of accurately determining their position in order to plan the route to their targets. The thesis presents an approach to solving the positioning problem in an environment where neither local odometry, global positioning nor image recognition individually provide sufficient data for accurate position determination. The effectiveness of several strategies for using each data source as a source of information on the agent's position and the positions of obstacles are evaluated, as is the effectiveness of different methods for combining information sources for increased accuracy. Based on simulation results, selected solutions are implemented and tested in real robots in a realistic testing environment.