2025 BrainChip Discussion, page-27

Behind door 2 of our is a topic that could unlock many future functionalities – .

Autonomous driving and the increased use of AI are just two examples that will lead to significantly higher energy requirements, making efficiency an even more crucial factor for vehicles in the future. This calls for innovative algorithms and hardware components that overcome current limits.

Together with different research partners, my team is breaking new ground in our research into artificial neural networks that mimic the workings of the human brain. The conventional von Neumann architecture based on processing binary signals supported by a separate memory bank is replaced by spiking neurons and synapses that use event-based processing and local memory. The former only sends signals when something actually happens (events), while the latter eliminates the bottleneck between processing and memory.

This could make AI computations significantly more energy-efficient and faster. For instance, current computers require up to 3000 watts to control sophisticated autonomous driving functions. Neuromorphic computing could do this in future with just 300 watts.

Our safety systems could, for example, recognise traffic signs, lanes and other road users much better and react faster, even in poor visibility. And they could do so 10 times more efficiently than current systems. There would be benefits in using a neuromorphic camera for interior monitoring, for example. Instead of full images (frames), it delivers individual pixels (events).
The process is extremely fast with minimal delay. This means, for instance, a rapid system reaction to the blinking of a driver’s eye caused by fatigue.

Our work in this field is still very much in the research phase, with much still to do. Nevertheless, we are encouraged and excited by the progress we are making.

Next week, I’ll open door 3, which takes us away from the intangible world of computing to something far more tactile.