複数のGNSSドップラー速度を複合した小型UAVの姿勢推定

抄録

This paper describes a small unmanned aerial vehicle (UAV) attitude estimation method using multiple global navigation satellite system (GNSS) doppler. In 3D measurement by a UAV with light detection and ranging (LiDAR), it is most important to estimate UAV attitude precisely. We have been developed a UAV with a LiDAR and six GNSS receivers to generate a 3D map. By using the developed method, we can estimate UAV attitude with 0.2 deg accuracy. And we also have been developed a UAV attitude estimation method with six GNSS receivers and inertial measurement unit (IMU). By using this method, we can estimate UAV attitude with 0.1 deg accuracy. However, in these previous researches, system configuration tends to be complicated by using IMU other than six GNSS receivers. Therefore, we now propose a UAV attitude estimation method with about same accuracy without using IMU by using only six GNSS receivers. In proposed method, we focus on GNSS doppler that we did not use in previous method. By calculating UAV angular velocity using multiple GNSS doppler and using it with Kalman Filter, we achieved 3D measurement in 55.4 mm horizontal Root Mean Square (RMS) error. In addition, by using six GNSS antenna positions that we did not use in the previous method with Extended Kalman Filter (EKF), we achieved 3D measurement in 54.3 mm horizontal RMS error.