A robot can often perform many tasks that people find incredibly difficult – meticulous number manipulation and data recall among them. But it can’t easily identify unknown objects from a random tray or bin and pick them up, until now. The MTC has made a breakthrough in intelligent robotics by developing a system that can teach itself how to pick up unknown objects, without the need for expensive sensors or lengthy programming.
The new robotic process uses a deep neural network, low-cost depth cameras and robotic arms with vacuum pick-up cups. The inexpensive components and highly flexible self-learning mean that the system can be put in place quickly and affordably, adapting to the individual needs of different industrial uses from picking components to cosmetics containers and even fruit. Working alongside a company’s workforce, this system is set to revolutionise item picking across all industrial sectors.