In this thesis, we investigated the use of the UR5e robot, the Visionary-T camera, and a conveyor belt for automated bread dough cutting. The goal of the research was to develop an efficient and accurate dough notching system that would increase productivity and reduce the need for manual labour in bakeries.
The system consists of three main components: the robot UR5e for precise slicing, camera Visionary-T for capturing bread point clouds, and a conveyor belt for transporting the bread through the process. We also used a computer to process the point clouds, using the Python programming language.
The system components are connected to the laboratory network via an Ethernet connection and communicate via TCP/IP protocol. The Visionary-T camera is connected to the UR5e robot. When the laser on the conveyor belt detects the presence of dough, the belt stops, and the camera captures a point cloud, which is then sent to the computer via the FTP protocol.
The computer processes the point cloud using an algorithm that includes statistical outlier removal and plane removal using Random Sample Consensus (RANSAC). We manually determine the plane coefficients based on the setup of the Visionary-T camera. The point cloud is then smoothed and the axes are determined using Principal Component Analysis (PCA). At this point, the dimensions of the dough are also measured and the optimal square-shaped box for the bread dough is designed. Depending on the user's requirements, the cutting angle and the number of cuts are determined, with the user specifying how the bread dough should be cut. Each specified point contains three coordinates (x, y, z), which are sent to the robot after proper calibration, since the camera and robot use different coordinate systems. For the purpose of this research, we have defined three cutting lines, which means 18 points (3 lines x 2 points x 3 coordinates).
When the robot receives these points, the belt restarts, and the robot starts cutting the bread dough. The system allows the cutting mode to be adjusted via the computer, allowing various shapes and cutting patterns of the bread dough.
The research was conducted experimentally by integrating and optimising the performance of all components. The main findings include increased accuracy of bread dough notching, reduced waste and increased production speed. The system is adaptable to different types and sizes of bread dough, further enhancing its usefulness in bakeries. Based on the results, we concluded that the system is reliable and efficient for use in bakeries.
|
