Tiny biomimetic devices are indispensable for human-inspired robotics and computing applications. A paradigm change from sensing to perception, aided by machine learning and deep neural networks, could revolutionize perceptive intelligence such as computer vision and voice processing.
It is well known that the human brain receives up to 80% of its information through the eyes, specifically through the rods and cones in the photoreceptor cells. The rods and cones absorb light, with the rods being found in much higher density levels than the cone cells.
In a new paper published in Light: Science and Applications, a research team at King Abdullah University of Science and Technology (KAUST) have investigated the creation of hybrid photoreceptors that could develop an efficient artificial retina network. The paper, entitled "A Flexible Capacitive Photoreceptor for the Biomimetic Retina," outlined the development of a composite material of perovskites and polymers. The author, Mani Teja Vijjapu, firmly believes that not everything around us can be felt or seen, hence the need for these sensors.
The research team focused on emulating the function of the rod cells. They are photosensitive due to the presence of certain pigments that are responsible for the change in action potential upon light illumination. Photoreceptor cells encode light information and transmit information to other cells for vision processing in the brain. The retina follows computing in the sensor methodology, so artificial retina networks are faster and more intelligent than conventional image processing devices.
To create the artificial retina network, the research team developed a light tunable artificial neuron that incorporates the functions of photoreceptors and ganglion cells. They opted for hybrid perovskites due to their exceptional optoelectronic properties. These properties include excellent light absorption, long carrier lifetime, low trap density, giant optical anisotropy, and high carrier mobility. They also incorporated polyvinylidene fluoride (PVDF) based ferroelectric polymers due to their high capacitance and charging & discharging efficiencies.
The researchers demonstrated a light-sensitive capacitive photoreceptor (CPR) that can mimic the retina's rod cells. Their hybrid materials are photosensitive and can be tuned to the dielectric properties. The photoresponse of the CPRs could be reproduced easily while also resisting the perovskites sensitivity to humidity and oxygen. The researchers also reported the longest stability of hybrid perovskites based devices at 129 weeks, increasing efficiency and cost-effectiveness.
The developed system has enormous potential in developing artificial retina networks and biomimetic eyes for perceptive intelligence applications. These CPRs could find a unique place in developing actuators for robotic applications. They could also provide opportunities for optoelectronic devices for optical communications.