Virtual Manufacturing is a supplier of lean-based production development services. We believe in and are willing to take on the challenge of combining technology, methods and hard work around production development in order to achieve results.
To ensure that decisions in a logistics project are based on facts positional tracking is of high importance. There are technologies available today that either require a long setup time or are very expensive. The signal-based tracking systems (Bluetooth, UWB, WiFi, etc.) are all sensitive to disturbances and require a long setup time. On the other hand, there are systems that are used for AGVs, one of which is Simultaneous Location and Mapping (SLAM).
We now want to investigate if visual SLAM can be utilized to provide an alternative solution to the challenge by investigating how machine learning based software would perform in the same application. We are interested in tracking tools, tolley, operators, vehicles, etc.
Proposed research questions
Research question 1: What is the accuracy we could get from Visual SLAM?
Research question 2: How does monoslam compare to stereoslam?
We would prefer that the development is done in an open source programming language. Thus, knowledge in programming in C++, C#, Python, or similar is crucial.
(Left)LIDAR based SLAM. (Right) Visual SLAM.
Number of students