A hybrid Kalman filter-fuzzy logic architecture for multisensor data fusion

In this work a novel hybrid Multi-Sensor Data Fusion (MSDF) architecture integrating Kalman filtering and fuzzy logic techniques is explored. The objective of the hybrid MSDF architecture is to obtain fused measurement data that determines the parameter being measured as precisely as possible. To reach this objective, first each measurement coming from each sensor is fed to a Fuzzy-adaptive Kalman Filter (FKF), thus there are n sensors and n FKFs working in parallel. The adaptation in each FKF is in the sense of adaptively adjusting the measurement noise covariance matrix R employing a fuzzy inference system (FIS) based on a covariance matching technique. Second, another FIS, here called a fuzzy logic observer (FLO), is monitoring the performance of each FKF. Based on the value of a variable called Degree of Matching (DoM) and the matrix R coming from each FKF, the FLO assigns a degree of confidence, a number on the interval [0, 1], to each one of the FKFs output. The degree of confidence indicates to what level each FKF output reflects the true value of the measurement. Finally, a defuzzificator obtains the fused estimated measurement based on the confidence values. To demonstrate the effectiveness and accuracy of this new hybrid MSDF architecture, an example with four noisy sensors is outlined. Different defuzzification methods are explored to select the best one for this particular application. The results show very good performance.