Intel helps robots recognise new objects

  • September 14, 2022
  • Steve Rogerson

Intel Labs has found a way to improve interactive continual learning for robots.

In collaboration with the Italian Institute of Technology and the Technical University of Munich, Intel has introduced an approach to neural network-based object learning.

It targets applications such as robotic assistants that interact with unconstrained environments, including in logistics, healthcare or elderly care. This research could help improve the capabilities of future assistive or manufacturing robots. It uses neuromorphic computing through interactive online object learning methods to enable robots to learn new objects after deployment.

Using these models, Intel and its collaborators demonstrated continual interactive learning on Intel’s neuromorphic research chip, Loihi, measuring up to 175 times lower energy to learn a new object instance with similar or better speed and accuracy compared with conventional methods running on a central processing unit (CPU).

To accomplish this, researchers implemented a spiking neural network architecture on Loihi that localised learning to a single layer of plastic synapses and accounted for different object views by recruiting new neurons on demand. This enabled the learning process to unfold autonomously while interacting with the user.

The research was published in the paper “Interactive continual learning for robots: a neuromorphic approach”, which was named best paper at this year’s International Conference on Neuromorphic Systems hosted by Oak Ridge National Laboratory.

“When a human learns a new object, they take a look, turn it around, ask what it is, and then they’re able to recognise it again in all kinds of settings and conditions instantaneously,” said Yulia Sandamirskaya, robotics research lead in Intel’s neuromorphic computing lab and senior author of the paper. “Our goal is to apply similar capabilities to future robots that work in interactive settings, enabling them to adapt to the unforeseen and work more naturally alongside humans. Our results with Loihi reinforce the value of neuromorphic computing for the future of robotics.”

In a simulated setup, a robot actively sensed objects by moving its eyes (event-based camera or dynamic vision sensor), generating microsaccades. The events collected were used to drive the spiking neural network on the Loihi chip. If the object or the view is new, its SNN representation is learned or updated. If the object is known, it is recognised by the network and respective feedback is given to the user.