MIE.62 – Revolutionizing Radiation Work: Simplified Control of Robot Arms with a Quick Learning Curve at Fermilab

• Jacob Chesney
• Christian Garcia
• Min Kim
• Jose Lopez
• Alyssa Mejia
• Briana Mendoza

Effects of radiation exposure on the human body have been studied for over a century, which can result in terminal illnesses, such as cancer or even death. The National Regulatory Commissions (NRC) has established occupational dose limits depending on the affected part of the body. The annual amount of dosage for the total body is 5,000 mrem in the United States. (National Regulatory Commission n.d.). Therefore, it is essential to establish these dose limits for radiation workers to prevent terminal illness. The studied impacts are influenced by radiation dosage, type of radiation, and exposure duration. This project is targeted toward company employees that experience radiation exposure. Considering this, all exposure to radioactive environments must be handled safely. The application of human-robot collaborative systems can be a beneficial tool in dangerous or harmful environments for human operators. This project’s scope involves the operator to efficiently use the industrial collaborative robotic arm (UR3e) in an irradiated environment to complete tasks while being able to operate it from a safe environment through an accurate and precise user-friendly device. To improve the safety of the operator, research was conducted on remote-controlled systems that would reduce exposure to the operator in hazardous environments. Collaborative robots also tend to be very difficult to operate and often have a difficult learning curve. To overcome this challenge the research team developed a user-friendly interface for the control system, which would have an easy learning curve for any operator. For the controller interface comparisons were made between designs of a joystick, glove, and hybrid technologies. Research into microcontrollers and microprocessors—such as Arduino, Raspberry Pi, and Nvidia Jetson—allowed the team to understand how the operator’s commands from the controller might be able to communicate with the robot system. A House of Quality and Function-Means Tree was used to help organize previously discussed customer needs, important metrics, and design concepts influenced by the research of existing human-robot collaboration systems. After conducting an analysis of controller systems, the team decided to design a controller in the form of a miniature-scale robot system. This controller uses encoders and decoders that translate signal pulses into positions and commands the UR3e can understand. A Raspberry pi is used to manage the communication and execution of commands between the controller and UR3e. These solutions effectively answer the established problems for the project. The control system is user-friendly and significantly reduces the learning curve, while also reducing hazardous environmental exposure for the operator Utilizing the developed control system, the manipulator can perform precise movements and reach the desired position with high accuracy. The controller is equipped with multiple switches. These are designed to grasp objects securely and rotate the gripper for higher precision and less demand of effort from the user. These functions allow the robot to execute more complicated tasks, such as screwing and unscrewing bolts while keeping it easy to control.