Machine Learning-enabled Biomimetic Electronic Olfaction Using Graphene Single-channel Sensors

Abstract

Olfaction is an evolutionary old sensory system, yet it provides sophisticated access to information about our surroundings. Inspired by the biological example, electronic noses (e-noses) in combination with efficient machine learning techniques aim to achieve similar performance and thus digitize the sense of smell. Despite the significant progress of e-noses, their development remains challenging due to the complex layout design of sensor arrays with a multitude of receptor types or sensor materials, and the need for high working temperature. In the current work, we present the discriminative recognition of odors utilizing graphene single-channel nanosensor-based electronic olfaction in conjunction with machine learning techniques. Multiple transient features extracted from the sensing response profile are employed to represent each odor and used as a fingerprint of odors. The developed electronic olfaction prototype exhibits excellent odor identification performance at room temperature, maximizing the obtained results from a single nanosensor. The developed platform may facilitate miniaturization of e-nose systems, digitization of odors, and distinction of volatile organic compounds (VOCs) in various emerging applications.