Robots and self-driving cars could soon benefit from a new kind of brain-inspired hardware that can allegedly detect movement and react faster than a human. A new study published in the journal Nature Communications details how an international team built their neuromorphic temporal-attention hardware system to speed up automated driving decisions.
The problem with current robotic vision and self-driving vehicles is a significant delay in processing what they see. While today’s top AI programs can recognize objects accurately, the calculations are so complex that they can take up to half a second to complete. That may not sound like a lot, but at highway speeds, even a one-second delay means a car travels 27 meters before it even begins to react. That is too long and too slow a reaction time.
Copying human vision
To solve this problem, the team worked on a hardware solution rather than tinkering with software, modeling it on how human vision works. When we view a situation, our visual system doesn’t analyze every detail at once. It first detects changes in brightness and movement, then processes the more complex details later.
The researchers built a chip that essentially does the same thing. It has a 4×4 array of specialized transistors that act as a filter. Instead of sending a whole video to the main computer, it identifies key changes in a scene. Because the computer only has to look at these regions rather than the entire image, the entire visual system runs faster.
In laboratory tests, this new system processed motion data four times faster than current state-of-the-art algorithms. Consequently, the hardware reduced visual processing time to around 150 milliseconds, roughly in the range of human visual perception.
The team also tested their technology on cars, drones, and a robotic arm. For the cars, the system improved perception and motion-related tasks by up to 213.5%, while it reduced visual processing time for drones, allowing for faster reaction and navigation decisions. And for the robotic arm, the system achieved up to a 740.9% improvement in grasping success rate for objects moving at high speeds.
“Our method outperforms state-of-the-art algorithms, achieving an average 4X improvement in processing speed while maintaining or enhancing accuracy in motion prediction, object tracking, and segmentation,” commented the researchers.
Following successful testing, the study authors want to move from a lab setup to a larger-scale version of the chip that can be integrated into self-driving cars and industrial robots. https://techxplore.com/news/2026-02-bio-chip-robots-cars-react.html
Recent Comments