In order to improve quality assurance in robotic casting, the data obtained from camera sensors helps to understand important process values. RTMaps (Real-Time Multisensor applications) plays to its strength to process and analyze this data.
The Intelligent and Skill Robotics Laboratory, Department of Mechanical Engineering, College of Science and Engineering, Aoyama Gakuin University, is conducting research on real-time flow measurement and robot control to improve product quality and yield in the robotic casting pouring process. The laboratory used RTMaps to exhibit its research at the International Robot Exhibition (iREX).
“Since molten metal cannot be flushed into the exhibition hall, we used wine to estimate the flow rate,” explains Professor Tasaki about the demonstration setup. The data from two cameras was input into RTMaps. With RTMaps, it was easy to recognize the width and curvature of the fluid and to estimate the flow rate. Basic image processing was performed using the image processing and OpenCV components preset in RTMaps. By using a Python bridge for the original logic, RTMaps can be used without wasting any of the research results created to date. In addition, sensor fusion when using multiple sensors can be easily achieved with RTMaps by simply configuring the components. Professor Tasaki states: “In addition to flow rate estimation, we also exhibited a demonstration of wine glasses being carried quickly and filled carefully according to the seated position of visitors. This demonstration was created in our laboratory in less than two weeks by students with no training.” The system used a camera to detect the visitor's position, position the wine glass with a robot arm at the designated coordinates, and pour the wine. RTMaps facilitated sensor input and sensor fusion, allowing the researchers to focus on their research subject.
Courtesy of Aoyama Gakuin University
dSPACE MAGAZINE, PUBLISHED MARCH 2024