Machine Learning-Based Web Service for Robot Programming with Cloud Sustenance

The utilization of robots is already commonplace in industrial automation. However, there is still ample opportunity for additional automation. An area that offers potential for enhancement is the adaptation of a manufacturing system to novel and unfamiliar items. Presently, this task entails new programming of the robotic device and the recalibration of the image recognition algorithms, should sensors be involved. Small and medium-sized organizations tend to avoid investing in the expertise of specialists due tothe required time and effort. Our proposal involves creating a digital replica of a physical manufacturing line, known as a digital twin. This digital twin will imitate the production system and collect labeled data. The data will be used to train a deep learning element. A trained artificial brain network will be used to identify and pinpoint the location of the Noticed items. This enables the production process to efficiently handle all familiar and unfamiliar products with more adaptability. The Artificial neural networks element is situated on a cloud infrastructure and can be reached via an internet service, enabling any staff member to commence the training, irrespective of their programming proficiency. To summarize, our method focuses on both increasing automation in production and utilizing synthesizing data to support artificial learning.