CAREER: Computational Models for Sensor-Based Machine Olfaction

Texas A&M Engineering Experiment Station, College Station TX

Investigators

Abstract

This is the first year of funding of a 4-year continuing award. The PI will focus on the development of a biologically plausible framework for sensor-based machine olfaction (SBMO), with an emphasis on computation, analysis and instrumentation. The specific objectives are: (1) To develop a computational architecture based on neuro-morphic models of the biological olfactory system in order to improve the signal-processing and cognitive capabilities of SBMO; (2) To validate the perceptual plausibility ofthe computational architecture through comparative studies with the "gold standard" for olfactory perception - human-panel sensory analysis; (3) To develop advanced sensor interrogation techniques in order to improve selectivity, sensitivity and robustness of commercial conductivity-based gas sensor arrays. The results will lay the foundations fora new generation of SBMO systems by helping bridge the gap between multivariate chemical sensing and human olfactory perception. Improved analytical capabilities, as a result of advances in both signal processing and sensor instrumentation, will broaden the range of applications for SBM0.