Radar Distance Positioning System : A Particle Filter Approach

Abstract: Positioning at sea has been important through all times. Thousands of years ago sea men used the stars to navigate. Today GPS is the most used positioning system at sea. In this thesis an alternative positioning method is described and evaluated. The advantage with the method is that it is independent of external systems which make it harder to interfere with than GPS. By calculating the distance to land using radar echoes (measured from the ship), and compare the distances to a digital sea chart a position can be estimated. There are several problems that have to be solved when using this method. The distance calculation and the comparison with the sea chart result in a non-linear system. One way to handle this non-linearity is the particle filter, which is used in this thesis. When using authentic radar data to estimate a position from an area of 784 km2, the system can isolate a small region around the correct position in two iterations. The system also manages to estimate the position with the same precision as GPS when the ship is moving.