Faster artificial vision, organelle therapy & ant injury treatment strategies
Your new Strategy Toolkit newsletter (March 24, 2026)
(1) The eyes have it…
So far, so good. Multiple cities around the world are growing ever more comfortable with automonous “driver-less” vehicles on their streets, whether it be the ubiquitous white Waymos in SF or the white Apollos in Shanghai.
But success inevitably leads to more complexity, and the technology’s inventors are turning once again to biology for innovating their product strategies. For example, improving the accurate vision and quick reflexes needed for every conceivable circumstance or surprise…
“Shuo Gao, a roboticist at Beihang University in China, wondered if biology might have the answer. Human eyes tame the complexity of the world by focusing attention only where it is needed. Central to this process is a region of the brain known as the lateral geniculate nucleus (LGN). The LGN acts as a relay station in the visual pathway, receiving information from the retina—where visual stimuli are converted into electrical signals—and passing it on to the brain’s visual cortex, where those signals are processed. But the LGN also plays an important filtering role, indicating to the visual cortex where processing power should be prioritised. Because the LGN’s filter is sensitive to changes in both time and space, it allows the brain to efficiently identify and track rapid movement, whether from a changing traffic light or a pedestrian crossing the street.
“Dr Gao and his team aimed to introduce an LGN-like layer into artificial vision systems to guide the attention of optical flow algorithms. Doing so with traditional computer chips, in which the circuits that process information are kept separate from those that store data, would not have given them the speed-up they needed. Instead, the researchers turned to so-called neuromorphic hardware, which mimics the human brain by having the processing and storage functions integrated into the same bit of circuitry.
“The researchers developed a novel piece of neuromorphic kit to imitate the LGN. Part of the device’s circuitry was designed to track changes in light intensity over time. This allows the device to build up a picture of where motion is occurring within a given environment and prioritise regions for optical-flow analysis.
“The researchers tested the new setup in a variety of contexts—including autonomous driving and robotic-arm operation—to see how it performed. The scientists found that their system operated at approximately four times the speed of existing optical-flow methods while maintaining or improving accuracy. Performance increases were particularly notable for autonomous driving, where the accuracy doubled. The system surpassed human-level speeds in most cases.”*
