Route programming on a mobile barn cleaner

The ‘Lely Discovery’ is an automated mobile barn cleaner, which routes are programmed by a ‘teaching method’. The routes are specified by a linear action sequence to be executed by the robot. Teaching a route to the Lely Discovery requires users to execute every action first by walking the robot through the barn. The route teaching method is time consuming and the current interface of the teaching method is not user-friendly. When a route requires changes for optimizing the route, the user has to re-execute all the route actions on the robot to and from the position where he wants to change the route. This thesis explores the implementation of a route drawing system which enables users to draw routes on a map. This system can generate an action sequence from the drawn routes on the map. Users can program routes 2 – 3 times faster than with the teaching method and route changes can be made in short notice. In this thesis the results of a user study are presented and the results of the tested routes on the robot via a first prototype of the drawing system. Although the results are not convincing, they provide useful leads for the design of a second prototype.