Multi-UAVs Multi-target Location Algorithm Based on Multi-frame Hypothesis Testing

Abstract

Infra-red Sensor is the commonly used to get the targets’ location for Unmanned Aerial Vehicle (UAV). For long distance target searching, Infra-red Search and Track (IRST) system passively achieves the bearing and elevation measurements of remote targets. Single UAV with IRST is difficult to locate the target in three dimension space (range, bearing, and elevation) because it can’t obtain any range information directly from the sensor measurements. In the paper, a sequential multi-frame hypothesis testing algorithm is presented for Multi-UAVs target location. Firstly, each single UAV obtains the targets’ angle-only track by utilizing the strong tracking filter (STF) and global nearest neighbor (GNN) angle-only tracking algorithm. Then, double UAV cross target locating track is achieved by recursive least square estimation algorithm. The Further UAV’s angle-only track is used for sequential multi-frame hypothesis testing with the target located positions of double UAV. During the certain M frame, if the N frame hypothesis testing is satisfied, the target track by double UAV is true target, and is updated by the further UAV angle-only track. On the contrary, the target track is false alarm target, and is deleted from the target track set. Consider the fluctuation of target track, a recursive least square algorithm is particularly used to smooth the target track. The simulation results show that the proposed algorithm can obtain effective performance for multi-UAV target location.