Long and Short Term Maneuver Trajectory Prediction of UCAV Based on Deep Learning

Lei Xie,Zhuoran Zhang,Zhenglei Wei,Andi Tang,Dali Ding

doi:10.1109/access.2021.3060783

Abstract

Autonomous air combat technology of unmanned combat air vehicles (UCAVs) is a hot issue that is currently being studied by various countries, and maneuvering trajectory prediction is an important part of autonomous air combat research. To address the difficulty of maintaining high prediction accuracy and short prediction time simultaneously in maneuvering trajectory prediction, this paper proposes a maneuvering trajectory prediction method that is based on a layered strategy, which combines long-term maneuvering unit prediction and short-term maneuvering trajectory prediction. In long-term maneuvering unit prediction, the complex trajectory is divided into 21 types of maneuvering units using the four characteristics of maneuvering trajectories, and a maneuvering unit library is established. On the basis of the deep echo state network(DeepESN), to capture multiscale prediction input parameters, autoencoder (AE) technology is incorporated. In addition, to increase the prediction accuracy, adaptive boosting (Ada) learning technology is utilized to build a strong predictor, and seven prediction networks are compared. The results demonstrate that the proposed method realizes the highest prediction accuracy. The single-step prediction time is about 0.002 s, which meets the time requirement. In short-term maneuvering trajectory prediction, the long and short-term memory (LSTM) network is analyzed, and the gaussian random walk strategy particle swarm optimization (GWSPSO) algorithm is used to update the internal weights and biases of the network to overcome the problems of “gradient disappearance” and “gradient explosion”, and a data sharing method is proposed for overcoming the no directionality of optimization algorithms. Compared with four traditional networks, the results demonstrate the method that is proposed in this paper performs better. Compared with the sampling time of 0.3 s, the short-term prediction time of 0.05 s can also meet the requirements. Finally, a long- and short-term layered prediction method is used on a group of complex maneuvering trajectories. The results demonstrate that the prediction accuracy is significantly increased and the real-time requirements are satisfied.

Highlights

With the continuous development of artificial intelligence technology, the intelligence and autonomy of unmanned combat air vehicles (UCAVs), which are represented by the American ‘‘loyal wingman’’, have been significantly improved, but the existing degree of intelligence is far from meeting practical requirements [1]; the autonomous air combatThe associate editor coordinating the review of this manuscript and approving it for publication was Cong Pu .technology of UCAVs is currently a hot issue that is being studied by various countries, and it has been a persistent research topic for decades [2]
To overcome the disadvantages of machine learning methods and solve the problems of low prediction accuracy and long prediction time under complex maneuvering trajectories, we have made the following original contributions in this article: 1. This paper proposes a maneuvering trajectory prediction method that is based on a layered strategy that combines longterm maneuvering unit prediction with short-term maneuvering trajectory prediction
To further evaluate the performance and robustness of the long and short-term maneuvering trajectory prediction method, the adaptive boosting (Ada)-AE-DeepESN long term prediction model is combined with the gaussian random walk strategy particle swarm optimization (GWSPSO)-long and short-term memory (LSTM) short-term prediction model, and the UCAV kinematics model is used to randomly generate a set of maneuvering trajectories for prediction

Summary

INTRODUCTION

With the continuous development of artificial intelligence technology, the intelligence and autonomy of unmanned combat air vehicles (UCAVs), which are represented by the American ‘‘loyal wingman’’, have been significantly improved, but the existing degree of intelligence is far from meeting practical requirements [1]; the autonomous air combat. Reference [16] proposes a prediction method that uses the HMM to model the movement trend of the flight based the historical trajectory, and uses the Gaussian mixture model to predict the aircraft speed vector, but the prediction error is large in the case of high-speed maneuvers. Reference [18] proposes a sequence-to-sequence deep long short-term memory network (SS-DLSTM) for trajectory prediction, which increases the accuracy and robustness of the prediction, but it is only applied to the terminal airspace of aircraft navigation, and the trajectory of the terminal airspace is relatively smooth. 4. The long- and short-term layered prediction method is used on a group of complex maneuvering trajectories, and the results demonstrate that the prediction accuracy is significantly increased and the real-time performance satisfies the requirements.

MANEUVERING TRAJECTORY PREDICTION METHOD BASED ON LAYERED STRATEGY

ADAPTIVE BOOSTING-AUTOENCODER-DEEP ECHO

LONG TERM MANEUVERING UNIT PREDICTION METHOD

GWSPSO OPTIMIZATION OF THE INTERNAL WEIGHTS AND BIASES

SHORT-TERM TRAJECTORY PREDICTION SIMULATION

CONCLUSION AND FUTURE DEVELOPMENT