6 DOF Robot Control with Object Detection Based on Deep Learning

Computer vision is an area of artificial intelligence that enables computers and systems to extract meaningful information from digital photos, videos, visual inputs and conduct actions or make suggestions based on that information. Robotics with applied computer vision method spans a wide range of fields. In the manufacturing, systems in production lines are continuously being upgraded to increase work efficiency. An ABB IRB1200 5–0.9 robot was chosen for product selection. With the task of controlling the actuator to carry out the task of picking the products in the tray and placing them on the conveyor belt, so as to ensure accuracy, and optimizing time. The article proposes a solution using computer vision algorithm integrated on the robot system, performing object detection, synchronizing camera and robot coordinate systems, and through the Robot Operating System (ROS) platform, controlling the robot to perform the task of picking and dropping objects.