Sensor modeling and fusion algorithms for NOx measures towards zero emissions vehicles

Measuring NOx concentration in exhaust gases of vehicles is essential for emissions monitoring and control. Commercial NOx sensors provide a measure influenced both by the temperature and by the presence of other substances to which the device is sensitive. This paper describes a technique to correct the measure obtained from the NOx sensor by exploiting additional compensating sensors. Thanks to accurate sensors models, it is possible to develop an algorithm that provides a precise estimation of NOx concentration using the data collected from all sensors. The algorithm is suitable to be implemented on a low-complexity microcontroller-based system, as it requires only a few memory resources and a limited computational power.