MIT makes AI sight-to-touch breakthrough


Researchers from the Computer Science and Artificial Intelligence Lab (CSAIL) at MIT have recently created an innovative system that could allow for advanced humanoid machine functionality. The revolutionary development implements a predictive artificial intelligence that would enable a robot to construct visual reference data based on touch input, and conversely, it could form physical environment parameters from data received through visual sensors.

Mapping out the future

It might sound complicated, but it’s just imitating what humans do on a daily basis. It looks at an object or a material and calculates what physical properties it might expect, such as the density or texture. “By looking at the scene, our model can imagine the feeling of touching a flat surface or a sharp edge,” according to Yunzhu Li, CSAIL PhD student. “By blindly touching around, our model can predict the interaction with the environment purely from tactile feelings. Bringing these two senses together could empower the robot and reduce the data we might need for tasks involving manipulating and grasping objects.”

The system enables tactile input to be converted into useful visual information to aid the robot as it maneuvers through its environment. The researchers used the design with a robot arm to help it decide the location of an object without any sight. The applications for such a system are endless. Automation is undoubtedly the way forward, and it’s these types of groundbreaking concepts that will act as the stepping stones to our soon-to-be fully autonomous future.

“This is the first method that can convincingly translate between visual and touch signals,” states postdoctoral Andrew Owens from the University of California at Berkeley. “Methods like this have the potential to be very useful for robotics, where you need to answer questions like ‘is this object hard or soft?’, or ‘if I lift this mug by its handle, how good will my grip be?’ This is a very challenging problem, since the signals are so different, and this model has demonstrated great capability.”


About Daniel Scheepers 207 Articles
I've always possessed a natural proclivity towards the art of writing. A strong passion and curiosity for life experience has given me diverse insight into varying sectors of the world. Opportunities to direct my talents are always welcome. Searching the web for interesting and factual news offers me a previously unimagined sense of fulfillment. When I have the chance, I'll be looking to get a Bachelor Degree of Communication.

Be the first to comment

Leave a Reply

Your email address will not be published.