Self-driving cars and advanced robots rely on cameras, sensors, and artificial intelligence to understand the world around them.
However, these systems often struggle when lighting conditions change quickly or when bright and dark areas appear at the same time.
A new study may offer a solution. Researchers have developed an artificial vision system inspired by the human eye that can adapt to changing light conditions in just seconds.
The technology could help future self-driving cars, robots, and other smart devices see more clearly and make better decisions in challenging environments.
The research, published in Nature Communications, centers on tiny electronic components called photomemristors. These devices can both detect light and store information, making them useful for next-generation vision systems.
Traditional camera sensors are usually designed for specific lighting conditions. While they may perform well in bright sunlight or darkness, they can struggle in situations where both light and dark areas exist at the same time.
For example, a self-driving car traveling at night may need to see a traffic signal while also dealing with the glare of headlights and the darkness of the surrounding environment.
According to the researchers, this is where the new technology offers an advantage.
The design was inspired by the way human eyes adapt to light. Inside the eye, specialized cells called rods and cones work together to help us see under different lighting conditions. Rod cells are particularly useful in dim environments, while cone cells help us distinguish colors and details in brighter settings.
The research team wanted to mimic this natural adaptation process.
To do this, they built photomemristors using two main materials. One is titanium oxide, which captures light and converts it into electrical signals. The other is a flexible, gel-like material called PEDOT:PSS.
When exposed to different lighting conditions, the device changes its water content. In darker environments, the material absorbs water. In brighter conditions, it releases water and dries out. This process automatically adjusts the device’s sensitivity to light without requiring external control.
As a result, the system can continuously adapt to changing conditions, much like the human eye.
The researchers tested the technology using different levels of ultraviolet light and found that it could accurately detect light intensity while maintaining stable performance under varying humidity levels.
To evaluate its practical usefulness, the team created a simple artificial vision system by combining a small array of photomemristors with a neural network, a type of artificial intelligence modeled after the brain.
The system was then asked to identify the letter “F” displayed against backgrounds with different brightness levels. After only seven training rounds, it correctly recognized the letter more than 95% of the time, even in mixed-light environments.
One of the most remarkable findings was the speed of adaptation. Human eyes can take 20 to 30 minutes to fully adjust when moving between bright and dark environments. The new photomemristors adapted much more quickly while still capturing detailed visual information.
The researchers believe the technology could eventually improve the vision systems used in self-driving vehicles, factory robots, and other autonomous machines. In the longer term, it might even contribute to advanced artificial vision systems designed to assist people with visual impairments.
By combining the adaptability of the human eye with the speed of modern electronics, the new technology could help machines see the world more naturally and reliably than ever before.
