Guang-Zhong Yang’s plenary talk at IROS 2016 touched on one of the most exciting applications of artificial intelligence – their use in surgical settings, where harmonizing human-robot interaction is critical. To do this, Prof. Yang spoke about his concept of “perceptual docking,” a method by which a “seamless transfer of knowledge between the human and robot” can take place. Of course, the more seamless this transfer is, the more precise the surgery.
In order to facilitate this, in his talk, he spoke about the underlying decision process of the operator and how human intention and skills can be learned in real-time by the robots. Some of his group’s research questions involve how much assistance is needed by the robots, how to reliably detect the response and how the human can offer help when it is required.
Because scientists are still relying on primitive learning techniques, it is important, Prof. Yang says, to understand the context of how the action or decision is being made. This is aided by the visual, tactile, and even olfactory senses. For example, using eye tracking (vision) provides a much more implicit way of tracking what you want to do, as it is extremely fast and powerful. But, Prof. Yang warns, looking doesn’t necessarily mean a person is paying attention so there need to be other methods such as real time analysis via smoke from scalpel.
Prof. Yang is the director and co-founder of the Hamlyn Centre for Robotic Surgery at Imperial College London in the UK. He is a fellow of the Royal Academy of Engineering and a recipient of the Royal Society Research Merit Award and listed in The Times Eureka “Top 100” in British Science.
IROS 2016 will run until Friday at the Daejeon Convention Center (DCC) in Daejeon, Korea.
