Improving the Accuracy of Static Relative GPS Positioning using Genetic Algorithm


Over the years, the Global Positioning System (GPS) has evolved to become an important navigational and positional system and is widely used across the world. The system promises high accuracies if the navigational signals transmitted by the GPS satellites are observed accurately. The modeling of a single point and relative point determination of user position includes pseudorange measurements. Taylor series is used to linearize the nonlinear model. Two methods are used to estimate the three dimensional user position: Recursive Least Square (RLS) method and continuous Genetic Algorithm (GA) method.Real data is used and results show that the GA enhances the estimation of user position more than the RLS by high error minimization and the minimum number of available satellites needed. RLS with three satellites give an error that exceeds the allowable limits, while GA gives an acceptable error. Relative positioning method is more accurate than the point positioning method for both RLS and GA methods..