Researchers from the National University of Singapore (NUS) announced on Wednesday that they are carrying out work aimed at giving robots a sense of touch through artificial skin.
The two researchers, who are also members of the Intel Neuromorphic Research Community (INRC), presented a study that demonstrates the promise of visual and tactile sensation, combined with Intel neuromorphic processing for robotics.
Most robots today are based solely on visual processing and do not have the sense of touch that humans have.
The researchers hope to change that by using artificial skin, which NUS believes will be able to detect touches more than 1.000 times faster than the human sensory nervous system. Artificial leather, NUS said, will also be able to recognize the shape, texture and hardness of objects "10 times faster than the eye".
The director of Intel Neuromorphic Computing Lab, Mike Davies, said the research provides a glimpse into the future of robotics where information is felt and processed in a fact-based way.
NUS said activating a human sense of touch in robotics could significantly improve the current Functionality, offering the example of robotic arms equipped with artificial leather that could easily adapt to changes in factory-made products, using the sense of touch to recognize and grasp unknown objects with the right pressure to prevent slipping.
Intel is helping researchers by providing a chip developed inside the robot to draw accurate conclusions based on real-time skin sensing data.
They also need an artificial one brain who can finally achieve the perception and learning that is another very crucial piece of the puzzle. Our unique demonstration of a dermal artificial intelligence system with neuromorphs chip such as the Intel Loihi is an important step forward in power efficiency and scalability.
Using Intel's Loihi neuromorphic research chip in their original experiment, the researchers used a robotic arm with artificial skin to read words written in braille, transferring data to Loihi via in cloud to turn hand-felt gadgets into a "semantic concept".
Intel said the Loihi achieved more than 92% accuracy in classifying Braille letters, while using 20 times less power than a standard Von Neumann processor.
Based on this work, the NUS team further improved the capabilities of robotic perception by combining vision and touch data with a neural network. To do this, they instructed a robot to sort various opaque containers containing different amounts of liquid, using sensory inputs from artificial skin and a camera.
The received sensor data was then sent to GPUs and Loihi for comparison processing capabilities. The researchers recorded that the combination of sight and touch, brought 10% greater accuracy in classifying objects compared to a system that uses only vision.