Fuzzy-logic based methodologies for mobile mapping: enhancing positioning accuracy of GPS/GNSS measurements

This paper shows an application of the fuzzy-logic approach to mobile mapping issues, presenting results of fuzzy processing of the C/A-code-based GPS measurements (standalone static positioning), whose accuracy depends essentially on the satellite observation geometry, expressed numerically by the GDOP (Geometrical Dilution Of Precision) factor, and on the Signal-to-Noise Ratio (SNR). A fuzzy controller, developed in the Matlab environment, is presented together with the envisaged rule base and the design steps (fuzzification, inference engine, defuzzification). A quality index is derived from the fuzzy-logic based data analysis, which allows us to select the “best” GPS observables to be processed in the positioning equations. Experimental results are presented and discussed from real data gathered by a commercial GPS receiver (the Garmin® GPS25-LV 12-channel receiver). Significant improvements of the positioning accuracy have been obtained with respect to the “raw” fixes provided by the receiver, showing the usefulness of the fuzzy-processed position fixes