Discontinuous maneuver trajectory prediction based on HOA-GRU method for the UAVs

With the rapid development of artificial intelligence technology, the intelligence and autonomy of Unmanned Aerial Vehicles (UAVs) have been significantly improved. Because the real trajectory data is often discontinuous and random, the current aircraft maneuver trajectory prediction methods are far from meeting the practical requirements of the autonomous air tasks. Especially, in order to occupy a better position rapidly where it is easier to attack the enemy, a fast and accurate maneuver trajectory prediction method for the UAVs is proposed in this paper. Firstly, the prediction model of aircraft maneuvering trajectory is built by extracting characteristic information from the historical trajectory. Aiming at the problem of slow optimization speed and easy to fall into local optimization, a global aircraft maneuver trajectory prediction method based on the Hummingbird Optimization Algorithm (HOA) and Gated Recurrent Unit (GRU) is proposed. Then, the implementation process of the maneuver trajectory prediction method based on the above HOA-GRU network for the UAVs is presented. Finally, the aircraft maneuver trajectory prediction method is applied to a simulation training system with the discontinuous and random air task data. The simulation results show that the proposed method can predict the maneuver trajectory of the UAVs with discontinuous data in real time with less error and less time.