Relative Humidity Estimation Based on Two Nested Kalman Filters with Bicubic Interpolation for Commercial Cultivation of Tropical Orchids

This paper aims to reduce the number of polymer dielectric-based humidity sensors used in the orchid greenhouse monitoring system by replacing some of them with a mathematical model. Therefore, this paper proposes an interpolation technique based on two nested Kalman filters with bicubic interpolation. Our objective is not only to accurately estimate the value at the location of interest but also to make it possible for practical usage. Thus, the computational time complexity is one of the criteria. The humidity values are estimated by an outer Kalman filter, of which its prediction is made based on another inner Kalman filter that fuses information obtained from surrounded sensors. The bicubic interpolation technique generates data used in the measurement update stage of the outer Kalman filter. Experimental results show that the proposed method can improve the interpolated data accuracy by 16.88%, compared with using interpolation techniques alone. This paper also discusses the possibility of applying the proposed scheme in improving the accuracy of the sensor used for a long time.