Skip to Main Content
This paper focusses on implementing an inertial navigation system (INS) on a graphics processor unit (GPU). The GPU allows simultaneous processing of large amounts of data and multiple use of the algorithm at the same time. Using the stream processing abilities of the GPU an investigation is launched into the simultaneous running of several navigation algorithms (INS's) with different sensor error models for each INS algorithm. The simultaneous processing of several INS systems allows for faster simulation times and data analysis. Aiding an error perturbed INS with a Kalman filter (KF) is considered when tuning the KF to obtain the optimally tuned KF from n amounts of KF's.