2023: Testing vision-based off-road navigation with geographical hints for autonomous mobile robots (operating in wild nature)
The goal for the project is to demonstrate robot navigation from point A to point B in an off-road environment using a possibly outdated map as a navigational hint. For testing the final result the public orienteering events are used, where the robot must follow the given orienteering track from start to finish, using only an orienteering map or satellite photo. In the first phase of the project a dataset is collected that is used to train two machine learning models: the first proposes new waypoints based on local sensor information, and the second uses the map to evaluate which of those waypoints leads to the final destination fastest. The second phase of the project implements the navigation system on a real robot.