Towards Robust Object Detection and Pose Estimation as a Service for Manufacturing lndustries

In his dissertation, Martin Rudorfer proposes a prototypic service-based framework for object detection and pose estimation in manufacturing industries. It combines the architectural style and computing capabilities of existing cloud vision APIs with specific, industry-relevant machine vision algorithms. It has the potential to make robotic applications more flexible and foster the automation of high-mix low-volume production.

---

The trend towards high-mix and low-volume production demands flexible and reconfigurable control for assembly systems. In less-structured environments, object detection and pose estimation is a key capability to enable industrial robotics applications such as grasping, handling and assembling. The integration and interconnectivity of such automation functions is fostered by Industry 4.0 through the adoption of service-based ecosystems.
The main objective of this thesis is to create a service-based framework for object detection and pose estimation in manufacturing environments. This could be a viable alternative to traditional machine vision systems such as smart cameras and embedded PCs, which are challenged by the high diversity and fast-paced progress in the field of object detection and pose estimation. We approach this problem in three steps: First, by designing a service-based framework that allows to handle all methods uniformly. Second, by examining the integration of three exemplary object detection and pose estimation methods, and third, by characterizing the strengths and weaknesses of the proposed solution compared to traditional machine vision systems.