Design and Simulation of 5-DOF Vision-Based Manipulator to Increase Radiation Safety for Industrial Cobalt-60 Irradiators

Robotics has proved its efficiency in nuclear and radiation fields. Computer vision is one of the advanced approaches used to enhance robotic efficiency. The current work investigates the possibility of using a vision-based controlled arm robot to collect the fallen hot Cobalt-60 capsules inside wet storage pool of industrial irradiator. A 5-DOF arm robot is designed and vision algorithms are established to pick the fallen capsules on the bottom surface of the storage pool, read the information printed on its edge (cap) and move it to a safe storage place. Two object detection approaches are studied; RGB-based filter and background subtraction technique. Vision algorithms and camera calibration are done using MATLAB/SIMULINK program. Robot arm forward and inverse kinematics are developed and programmed using an embedded microcontroller system. Experiments show the validity of the proposed system and prove its success. The collecting process will be done without interference of operators, hence radiation safety will be increased.