Accuracy analysis of camerainertial sensor based ball-trajectory prediction

The RoboCup initiative formulated its long-term goal as winning against the most recent champion of the World Cup in 2050 with a team of autonomous robotic soccer players. At this point of time, the question if the initiative will achieve this goal cannot be an-swered. Although the progress within the robotic soccer community is remarkable, it is not known whether it will lead to competitive robotic soccer players until then or not. Legitimately, one may also ask the question the other way around: Is there an area of research, in which enough knowledge has already been obtained for the 2050 scenario? This thesis tries to give an answer to the question with respect to the sensing subsystem within the domain of soccer playing robots. More precisely, this thesis will reveal if the accuracy of ball trajectory prediction is sufficient for a hypothetical robotic soccer player under certain constraints. The sensor setup for this hypothetical robot is a camera-inertial sensor which allows per-ception of the environment and tracking of the sensor’s motion relative to itself. To an-swer the question if the accuracy is sufficient or not, an experimental-based analysis i