UVA Trajectory Prediction Model and Simulation Based on Bi-LSTM

The traditional trajectory prediction models have the problems of large model simplification and less consideration. Combined with the characteristics of flight trajectory continuity, time series and interactivity, a trajectory prediction model based on bidirectional long short term memory(Bi-LSTM) neural network is proposed. The position, heading, pitch, roll and relative information of the intruder UAV are simultaneously used as the input of the trajectory prediction model, which is more in line with the true trajectory change law. The established Bi-LSTM based trajectory prediction model is trained with adaptive learning rate learning algorithm considering momentum and speed, and performed with simulation contrastive analysis with trajectory prediction model based on Elman neural network. The results show that, compared with the trajectory prediction model based on Elman neural network, the average absolute error of the proposed model predicted by 200 points in different directions is less than 4 m, the 3D prediction effect is better, and can perform the trajectory prediction more accurately.