Design Of Cascaded Mimu/Gps Fusion Navigation Algorithm For Unmanned Helicopter

Autonomous unmanned helicopter requires avionics system to maintain stable attitude and to follow desired flight path. A cascaded fusion algorithm is designed in this paper to provide full flight information for small size of helicopter. It consists of two kalman filters to process MIMU, GPS and other sensors data in real time based on the characteristic of onboard sensors. The first filter estimates velocity and attitude via MIMU data, GPS velocity and earth magnetic field strength. The second kalman filter provides position through GPS position, barometer information and the first filter's calibrated velocity. Simulation shows that the proposed algorithm can estate the flight information with high accuracy and bandwidth. A navigation avionics system is implemented based on the proposed algorithm, and real flight test demonstrates that the algorithm can work well with real sensor data on helicopter.