There are different "levels" of nuclear waste, but none of them should be handled in person. This is why remote-controlled robots have become the standard tool for decommissioning nuclear facilities and processing radioactive materials. However, it is difficult to control every movement of a robot during complex tasks such as cleaning a nuclear reactor. That's why a team from Lancaster University has developed a semi-autonomous robot This could make the process faster and easier.
It is highly unlikely that a truly autonomous robot will be assigned nuclear dismantling tasks in the near future. After all, AI is still far from perfect, and the stakes are as high as they are when you handle high-level radioactive material in sufficient quantities to cause uncontrollable nuclear reactions.
The Lancaster robot slice the difference by adding intelligent skills to the process while leaving the human operator in the driver's seat. The team has created an imaging software allowing the robot to "see" the surrounding world and identify objects such as pipes, handles and other common materials in nuclear dismantling sites. It does this with the help of a Microsoft Kinect camera. Yes, the game accessory doomed to endure among those who need a cheap depth detection camera.
The robot has seizures that a human could clumsily control with the help of a joystick, but operators working with the Lancaster robot simply need to tell you what task to undertake thereafter. During testing, it took only four mouse clicks for an operator to point the robot at an object and select an action. It is up to the robot to determine the specificities of lifting, traction and cutting. The team says its semi-autonomous robot has largely outperformed manual operation.
This technology could speed up decommissioning work and speed up the training of new operators. However, the test was far from authentic. It took place in a laboratory that was devoid of radioactive material – any radiation would have been a problem for the robot, without industrial radiological protection. The robot has successfully completed tasks such as cutting plastic pipes, work similar to the one he would have to perform in the field. The researchers intend to continue to improve the design and operation. Future versions may be able to survive radiation exposure and relay data such as temperature and audio from the hot zone.