Estimation of UAVs' Attitude by using Multi-Sensor Integration

H.P. Thien, T. Mulyanto, H. Muhammad (Indonesia), and S. Suzuki (Japan)


State Estimation, Data Fusion, Filtering, Kalman Filter


As a major step toward the establishment of mathematical model as well as control development for an Unmanned Aerial Vehicle (UAV), sensing of vehicle state should be done in priority. Acceleration, velocity, attitude, and position of the vehicle should be determined in a direct or indirect way. In fact, these states could not be provided by an individual sensor. Hence, integration of multi-sensor is required. And, merging of sensors’ output is necessary to achieve accurate information. In this paper, the estimation of attitude for an UAV by fusing of multi sensor is dealt with. Two Kalman filters are used to merge the angular rate, acceleration from Inertial Measurement Unit and attitude data from Digital Magnetic Compass to obtain the effective information of attitude. Testing on developed sensory system shows that the fusion algorithm works rather well, and can be used to determine the attitude based on data come from several sensors at different update rate.

Important Links:

Go Back