GNSS error simulator for farm machinery navigation development

Display Omitted A new tool to support the development of sensor fusion in steering and navigation automation.Allan variance as a method for revealing dynamic positioning error characteristics.Spatial autocorrelation effect on the local GPS error.Accuracy results from dynamic GPS tests in farming conditions. Satellite positioning systems are widely used for determining the position of moving machines and vehicles. Different methods for improving the accuracy and reliability are often used by fusing dead reckoning systems to the satellite positioning. To develop these fusions and to determine sufficient positioning accuracies, a realistic simulator for positioning data is required. In this study, we developed a GNSS (Global Navigation Satellite System) error simulator to support the development of navigation capabilities. The simulator was based on real dynamic tests driven with an agricultural tractor. The driving speed of the tests was approximately 9km/h and the GNSS measurement rate was 10Hz. We used Allan variance calculations to determine the characteristics of the errors and Geary's C to determine the spatial autocorrelation of the error in agricultural fields. With these methods, we constructed a feasible error simulator capable of simulating positioning errors and positioning reliability for all typical GNSS positioning accuracy levels. The simulator uses White noise, Random walk, Flicker noise, Rate ramp, a First order Gauss-Markov process and the spatial autocorrelation map to generate errors for horizontal coordinates.