Intel breaking new ground in neuromorphics

Intel’s newly announced Loihi 2 neuromorphic research chip could break new ground in autonomous vehicles, according to Michael Milford, Professor at Queensland University of Technology’s Joint Centre for Robotics.

Intel last week announced the second generation of its Loihi product line, along with Lava, an open-source software framework for developing neuro-inspired applications.

Neuromorphic computing draws insights from neuroscience to create chips that function more like the biological brain. The second-generation chipset enables up to 10 times faster processing and up to 15 times greater resource density than its predecessor.

Professor Milford said the enhancements to the chipset promise new developments in the field of simultaneous localisation and mapping (SLAM), which is at the heart of the operation of autonomous mobile vehicles in unknown, GPS-deprived or dynamic environments.

“Visual place recognition is a particularly challenging part of SLAM that requires robust recognition and discrimination of hundreds of thousands of locations in different conditions,” he said.

Professor Milford said there is a wealth of inspiration to be derived from the navigation capabilities of animals in the field of neurotrophic computing

“Animals display amazing navigation capabilities, solving this large-scale memory formation and recognition problem with an unprecedented efficiency, flexibility and robustness,” he said.

“New features of Loihi 2 enable more complex neuronal models and neural plasticity rules, making it well-positioned to implement the next generation of biologically inspired navigation and map formation algorithms that could surpass today’s state of the art in the field.”

Image credit: ©stock.adobe.com/au/peshkova