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
Proceedings of the IEEE Sensors Conference, October 28-31, 2012, Taipei, Taiwan.
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
Download:
Paper: [PDF (1.7MB)]
@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}
}