New autonomous car vision system is 4 times faster than human brain

A multinational research team from China, Britain, Hong Kong, Saudi Arabia, and the United States has developed a novel hardware-based vision system for autonomous machines that reacts to hazards up to four times faster than the human brain. Traditional autonomous vehicles typically take about 0.5 seconds to respond to obstacles at 50 mph, traveling roughly 43 feet before braking, whereas humans react in about 0.15 seconds. The new system addresses this critical safety gap by implementing a “reflex” mechanism inspired by human vision, which prioritizes detecting sudden motion or changes before detailed scene analysis. Central to the innovation is a two-dimensional synaptic transistor array that filters out irrelevant visual data and detects motion changes within 100 microseconds, significantly accelerating reaction times without compromising accuracy. The system operates by capturing only moving objects and passing these signals to conventional computer vision algorithms for further processing, resulting in a speed increase over ten times that of traditional image processing methods. Laboratory tests demonstrated a 213.5% improvement in