Towards drift correction in chemical sensors using an evolutionary strategy

Gas chemical sensors are strongly affected by the so-called drift, i.e., changes in sensors’ response caused by poisoning and aging that may significantly spoil the measures gathered. The paper presents a mechanism able to correct drift, that is: delivering a correct unbiased fingerprint to the end user. The proposed system exploits a state-of-the-art evolutionary strategy to iteratively tweak the coefficients of a linear transformation. The system operates continuously. The optimal correction strategy is learnt without a-priori models or other hypothesis on the behavior of physical-chemical sensors. Experimental results demonstrate the efficacy of the approach on a real problem 1 . Categories and Subject Descriptors I.2 [Computing Methodologies]: Artificial Intelligence; I.5.4 [Computing Methodologies]: Pattern RecognitionApplications General Terms Algorithms Keywords Drift correction, Artificial olfaction, Evolutionary Strategies

• Type: Conference
• Year: 2010
• Where: GECCO
• Authors: Stephano Di Carlo, Ernesto Sánchez, Alberto Scionti, Giovanni Squillero, Alberto Paolo Tonda, Matteo Falasconi
• Keywords: Correct Unbiased fingerprint, Gas Chemical Sensors, GECCO 2010, Keywords Drift Correction, Optimization