Delta robots are used for applications where fast operations are required. Developed in cooperation with the industrial partners, this five-axis Delta robot in Testbed enables synchronization of its movements with the movement of the pallets on the conveyor. Researchers designed and fine-tuned a kinematic and dynamic model to compare the behavior of a real robot with the expected ones and to identify differences through machine learning algorithms.
This serves as a basis for machine operation diagnostics, so-called condition monitoring. The dynamic model is used for feedback control based on the force exerted on the robot’s end effector. The robot operates on an industrial 5G network and enables fast transmission of large amounts of data. Therefore, the integration of a neural network and computer vision could be implemented in the robot system. 5G SA communication is used to transfer the image from the camera to the server (edge device) and transmit the necessary position data back to the robot in real time. The experiment also uses augmented reality for predictive maintenance using a HoloLens kit connected via campus 5G SA network.
Delta Robot Demonstrator
The robot can be equipped with a handle to perform what is called a “manual guide” to learn the trajectories to be performed during normal operations. It operates on an industrial 5G network and enables fast transmission of large amounts of data. Therefore, the integration of a neural network and computer vision could be implemented in the robot system.
Two linear motor conveyors are integrated into the workstation. This design allows the robot to perform typical pick-and-place movements often used in the automotive industry.
This technology allowed the development of a Delta handheld guidance controller of relatively small size and with strong computing power executed on a server.
The second concept implemented is a learning mode, where the operator can demonstrate the desired movements to the robot by navigating with the handheld guidance controller and the neural network will help the operator to refine it.
The Delta robot is an example of a multi-axis positioning system with synchronized axes of motion. Each of the two machines is controlled by a separate control system, and the two controllers are interconnected to synchronize and interact with each other.
The communication module through the 5G SA network uses the computing resources on the server (edge cloud) for image processing by artificial intelligence algorithms, which helps in manual guidance to the desired position. In addition to the robot, a detailed kinematic and dynamic model has been created to continuously monitor the robot’s behavior and detect any deviations from the expected behavior.
