Javier González Monroy, Achim J. Lilienthal, Jose Luis Blanco, Javier González Jimenez and Marco Trincavelli
Calibration of MOX gas sensors in open sampling systems based on Gaussian Processes
Abstract:
Calibration of metal oxide (MOX) gas sensor for continuous monitoring is a complex problem due to the highly
dynamic characteristics of the gas sensor signal when exposed to natural environment (Open Sampling System - OSS). This
work presents a probabilistic approach to the calibration of a MOX gas sensor based on Gaussian Processes (GP). The proposed
approach estimates for every sensor measurement a probability distribution of the gas concentration. This enables the calculation
of confidence intervals for the predicted concentrations. This is particularly important since exact calibration is hard to obtain
due to the chaotic nature that dominates gas dispersal. The proposed approach has been tested with an experimental setup
where an array of MOX sensors and a Photo Ionization Detector (PID) are placed downwind w.r.t. the gas source. The PID is used
to obtain ground truth concentration. Comparison with standard calibration methods demonstrates the advantage of the proposed approach.
Keywords: Calibration, Estimation, Gas detectors, Microwave integrated circuits, Robot sensing systems, Training, Uncertainty
@INPROCEEDINGS{Trincavelli_etal:IEEESensors12:2012,
AUTHOR = {González Monroy, Javier and Lilienthal, Achim and Blanco, Jose Luis and González Jimenez, Javier and Trincavelli, Marco},
TITLE = {Calibration of MOX gas sensors in open sampling systems based on Gaussian Processes},
BOOLTITLE = {Sensors, 2012 IEEE},
VOLUME = {},
YEAR = {2012},
DATE = {October 28-31},
ADRESS = {Taipei, Taiwan},
PAGES = {1-4}
DOI = {10.1109/ICSENS.2012.6411464}
}