Multi-UAV Collaboration and IMU Fusion Localization Method in Partial GNSS-Denied Scenarios

Abstract

When it comes to disaster rescue, quickly and accurately locating the trapped people will not miss the golden rescue time, thereby improving the success rate of rescue. The existing research has focused on the use of unmanned aerial vehicles (UAVs) to locate stationary user equipment (UE). However, the simultaneous movement of the UAVs and the UE brings about problems of UAV coordinate deviation and range interference, resulting in poor accuracy and stability of UE localization. To solve the mentioned issues above, a fusion localization method based on multi-UAV collaborative ranging and user-side inertial measurement unit (IMU) module was proposed for GNSS-denied scenarios. First, the initial coordinate deviations of the UAVs were corrected by multi-UAV collaboration; Then, a pre-processing scheme based on density peaks clustering (DPC) was proposed to eliminate the abnormal interference in the ranging data; Finally, the unscented Kalman filter (UKF) was used to realize the fusion localization based on multi-UAV ranging and user-side IMU module. The simulation results showed that the proposed scheme can maintain the accuracy and stability of UE localization in the presence of ranging interference, and the positioning error was reduced by 21.4% compared with the commonly used extended Kalman filter (EKF). Moreover, the proposed localization scheme had a better performance in low-altitude rather than high-altitude flight scenarios.INDEX TERMS Density peaks clustering (DPC), fusion localization, inertial measurement unit (IMU), unmanned aerial vehicle (UAV), unscented Kalman filter (UKF)