“The leading semiconductor components distributor dedicated to the Asia-Pacific market, Dalianda Holdings, announced that its subsidiary Youshang has launched a 3D object gripping system solution based on Intel (Intel) OpenVino and Realsense Camera.
On June 2, 2021, Dalian General Holdings, a leading semiconductor component distributor dedicated to the Asia-Pacific market, announced that its subsidiary Youshang has launched a 3D object gripping system solution based on Intel (Intel) OpenVino and Realsense Camera.
Figure 1 – The Display board of Dalian Dayou Shang’s 3D object gripping system solution based on Intel products
In the manufacturing, metal processing, and food industries, there are still many loading and unloading jobs that rely on manual labor. However, due to the increasing lack of labor, the need for robots to automate loading and unloading gradually emerges. One of the difficulties in realizing automatic loading and unloading is that in many scenarios, workpieces such as components, metal blanks, and food packaging are scattered in containers. Compared to humans, it is difficult for robots to extract randomly placed parts from containers and place them precisely into machines. The 3D object clamping system solution launched by Dalian Dayou Shang can achieve precise object clamping through a series of advanced technological innovations, effectively solving the above problems.
Figure 2 – Dalian Dayoushang launched a 3D object gripping system solution based on Intel products. Scenario application diagram
The solution records 3D image data of objects through Intel Realsense D415 Camera, and sends the image data to Edge AI System through USB. The Edge AI System with built-in Intel OpenVino toolkit recognizes the relevant information of object position and attitude through image analysis and deep learning algorithm (X, Y, Z, Rx, Ry, Rz), the AI system can also upload the identified object type, status and other information to the cloud or local terminal and display the object information through the dashboard. The robotic arm and the Edge AI System communicate with each other through the TCP/IP protocol, and execute the obtained object position and attitude information (X, Y, Z, Rx, Ry, Rz) to grasp objects and realize the rapid deployment of AI recognition solutions.
Figure 3 – Dalian Dayou Shang launched a 3D object gripping system solution block diagram based on Intel products
Core technical advantages:
1.Intel RealSense D415 camera
• Low-cost 3D binocular depth camera;
• Provide a complete SDK for quick integration with the system;
• It can scan quickly and provide point cloud information;
• Independent innovation functions can be developed through ROS integration;
• Intelligent 3D object recognition.
2. Intel OpenVino Toolkit
• The trained model can be optimized;
• Support training frameworks commonly used by the industry and academia;
• Can be quickly deployed to Intel hardware platforms such as CPU, GPU, VPU, FPGA;
• Provide common pre-training models such as SSD, YOLO, etc.;
• Provide application examples of C++ and Python to shorten the program development cycle.
3. 3D object gripping system solution
• Automatically grip and place objects;
• Customized object recognition (models can be retrained according to customer needs);
• Transmission of object gripping information via standard TCP/IP interface.
• 3D camera: Intel RealSense D415 Camera;
• Operating system: UBuntu 16.04;
• Intel NUC Rugged Board with Core i3/i5 Processor;
• AI Inference Kit: Intel OpenVino Toolkit 2020.03;
• Memory: 4GB or more;
• Transmission interface: USB 3.0 and TCP/IP interface;
• Robot arm: 6-axis arm, 4-axis arm with TCP/IP interface;
• Accelerator Card: Intel® Movidius™ Myriad™ X Edge AI Module VEGA-320-01A1.