Artificial Olfactory Sensor Systems via a Seamless Neuromorphic Architecture

Neuromorphic olfactory systems have gained significant attention in recent years due to their promising applications in electronic noses, robotics, and neuromorphic data processing. Conventional gas sensors, while effective in detecting hazardous gas levels, often lack synaptic functions such as memory and recognition of gas accumulation which are key features needed for mimicking human-like neuromorphic sensory systems. In this study, we present a seamless architecture for a neuromorphic olfactory system capable of both detecting and memorizing nitrogen dioxide (NO2) levels during continuous exposure. The system autonomously triggers a self-alarm after 147 and 85 seconds at continuous exposure of 20 ppm and 40 ppm, respectively. Using thin-film transistor gas sensors with carbon nanotube semiconductors, NO2 detection occurs via carrier trapping, which exhibits long-term retention, making it suitable for neuromorphic excitatory applications. Additionally, the system's neuromorphic inhibitory function is demonstrated through gas desorption, regulated by programmable ultraviolet light exposure, achieving homeostasis recovery. These findings offer a promising pathway for developing an artificial olfactory system with advanced biological synaptic functions within a streamlined architecture