In the assembly processes of the consumer goods and general industries, the problems of missing parts and misassembly have always troubled manufacturing enterprises. DaoAI 3D robot vision technology of WeLinkirt provides an effective solution to this difficult problem.
User scenario: An assembly line of a leading consumer goods manufacturer, mainly producing various small household appliances. In the product assembly process, multiple components need to be accurately assembled onto the product body. The detection objects include small parts such as screws, capacitors, and chips to ensure that there are no missing parts or misassembly during the assembly process.
Pain points: In the current market trend, the market potential of automatic charging robot solutions for household and public scenarios is huge, which puts forward higher requirements for the production accuracy and quality of consumer goods. The assembly process of this manufacturer has problems such as low efficiency of manual inspection and high missed detection rate, and the labor cost remains high. The missed detection rate of manual inspection reaches 4%, the false alarm rate is 7%, and the changeover time is as long as 30 minutes, which seriously affects the efficiency of the production line and the product quality, and it is difficult to meet the market demand for high-quality products.
Technical Principles
DaoAI 3D robot vision uses a self-developed 3D camera for imaging, which can obtain the three-dimensional morphology information of objects. In the detection of missing parts or misassembly in assembly, the 6D pose estimation algorithm is used to accurately calculate the spatial position and attitude of each component. By comparing with the pre-set standard model, it can judge whether there are missing parts or misassembly. The reason why this method is effective is that the rich three-dimensional information provided by the 3D camera can reflect the object features more comprehensively, and the 6D pose estimation algorithm can accurately capture the subtle changes of components, which greatly improves the detection accuracy.
- 3D camera: Obtain high-precision three-dimensional morphology data to provide a basis for subsequent analysis.
- 6D pose estimation algorithm: Accurately calculate the position and attitude of components, achieving sub-millimeter hand-eye coordination.
- Transfer learning: Use existing models to quickly adapt to new products and detection tasks, reducing model training time.
- Brain - eye-body closed-loop: Realize real-time interaction between the camera, algorithm, and robot to ensure the efficient progress of detection and assembly.
WeLinkirt Solution and Product
Centered on DaoAI 3D robot vision, this solution deploys a self-developed 3D camera in the production line to collect 3D images of components in real-time. Through the 6D pose estimation algorithm, the images are analyzed to quickly determine whether there are missing parts or misassembly. When a problem is detected, the system will immediately issue an alarm and make timely corrections through the robot. At the same time, the transfer learning technology enables the system to quickly adapt to new product changeovers, and the model can be updated with only a small number of samples. The supporting DaoAI World model provides a unified base for cross-scenario generalization and continuous learning, further enhancing the performance and adaptability of the system.
DaoAI 3D robot vision achieves accurate detection of missing parts in assembly with advanced technology.
Quantifiable results: After adopting this solution, the missed detection rate is reduced to <0.6%, the false alarm rate is reduced by -63%, and the changeover time is shortened to 5 minutes. These significant results not only improve the efficiency of the production line and product quality but also reduce the production cost, making the enterprise more competitive in the market.
FAQ
What types of components can DaoAI 3D robot vision detect?
DaoAI 3D robot vision can detect small parts such as screws, capacitors, and chips. It is suitable for various components in the assembly processes of the consumer goods and general industries, achieving accurate detection through 3D imaging and 6D pose estimation.
What is the role of transfer learning in this solution?
Transfer learning enables the system to quickly adapt to new products and detection tasks by using existing models. The model can be updated with only a small number of samples, greatly shortening the changeover time and improving detection efficiency.
How is the detection accuracy of the system ensured?
High - precision three-dimensional morphology data is obtained through the self-developed 3D camera. Combined with the 6D pose estimation algorithm to achieve sub-millimeter hand-eye coordination, it can accurately capture the subtle changes of components and ensure detection accuracy.