Abstract
This master thesis deals with the selection of a suitable tracking solution for an “augmented reality picking system”. The existing system is able to navigate the picker through a warehouse and to guide him to the right storage locations. The new tracking solution shall support the picker while putting a box on the pallet by augmenting the correct positon. Such a tracking system has to be capable of tracking the pose (position and orientation) of the persons head in real time. For precise augmentation it has to fulfil high accuracy requirements. To verify the storage position of a box on the pallet, the picking system needs to be linked with an automatic control system. A prototype of such a system has been developed. The theoretical part of this thesis describes those tracking technologies, which qualify for the described task from a technical point of view. In cooperation with various tracking system providers, concrete solution scenarios were developed. These scenarios were evaluated using the “Analytical Hierarchy Process” and a cost-benefit analysis. For developing the prototype of the control system, which checks the storage position of a box on the pallet, a depth image camera (Microsoft Kinect) was used. The developed algorithm for identifying a box and checking its position is described. Furthermore test results on the measurement accuracy are reported.
Translated title of the contribution | Tracking variants for augmented reality picking systems |
---|---|
Original language | German |
Qualification | Dipl.-Ing. |
Supervisors/Advisors |
|
Award date | 24 Oct 2014 |
Publication status | Published - 2014 |
Bibliographical note
embargoed until 10-09-2016Keywords
- tracking
- augmented reality
- picking