Trajectory Prediction And Error Compensation Of UAV

In recent years,more and more light and small unmanned aerial vehicles(uavs)play an increasingly important role in maintaining social order,participating in rescue events and promoting economic development and other livelihood fields.The wide use of uavs in civil fields inevitably poses a higher challenge to their safety.The flight path deviation problem of unmanned aerial vehicle(UAV)is one of the most important research focuses.The main research work of this paper includes the following three parts:(1)A modularization-based hierarchical software architecture for UAV system is proposed to overcome that shortcoming of strong coupling and weak extension in the existing software architecture.The whole UAV system is divided into four layers,and each layer is divided into modules according to functions.At the same time,the function of UAV ground control station based on the improved architecture is increased and optimized.(2)Aiming at the problem of UAV’s deviation from flight path,a trajectory prediction algorithm based on long and short memory neural network and extended Kalman filter is proposed to predict the trajectory deviation of UAV so as to intervene in advance.By comparing the performance of the filtering algorithm,the neural network algorithm and the trajectory prediction algorithm proposed in this paper on the simulation experimental data and the real experimental data,the results show that the algorithm proposed in this paper can predict the trajectory for a long time and maintain a high prediction accuracy.(3)that double Kalman filt algorithm based on prediction is use to compensate the UAV which is predicted to deviate,and the comparison of the extended Kalman filter algorithm and the recursive average filt algorithm shows that the proposed algorithm has better compensation effect.