Abstract
Trajectory prediction plays an important role in modern air combat. Aiming at the large degree of modern simplification, low prediction accuracy, poor authenticity and reliability of data sample in traditional methods, a trajectory prediction method based on HPSO-TPFENN neural network is established by combining with the characteristics of trajectory with time continuity. The time profit factor was introduced into the target function of Elman neural network, and the parameters of improved Elman neural network are optimized by using the hybrid particle swarm optimization algorithm (HPSO), and the HPSO-TPFENN neural network was constructed. An independent prediction method of three-dimensional coordinates is firstly proposed, and the trajectory prediction data sample including course angle and pitch angle is constructed by using true combat data selected in the air combat maneuvering instrument (ACMI), and the trajectory prediction model based on HPSO-TPFENN neural network is established. The precision and real-time performance of trajectory prediction model are analyzed through the simulation experiment, and the results show that the relative error in different direction is below 1%, and it takes about 42ms approximately to complete 595 consecutive prediction, indicating that the present model can accurately and quickly predict the trajectory of the target aircraft.
Highlights
Aiming at the large degree of modern simplification, low prediction accuracy, poor authenticity and reliability of data sample in traditional methods, a trajectory prediction method based on HPSO⁃TPFENN neural network is established by combining with the charac⁃ teristics of trajectory with time continuity
The time profit factor was introduced into the target function of Elman neural network, and the parameters of improved Elman neural network are optimized by using the hybrid particle swarm optimization algorithm ( HPSO), and the HPSO⁃TPFENN neural network was constructed
An independent prediction method of three⁃dimensional coordinates is firstly proposed, and the trajectory prediction data sample in⁃ cluding course angle and pitch angle is constructed by using true combat data selected in the air combat maneuve⁃ ring instrument ( ACMI), and the trajectory prediction model based on HPSO⁃TPFENN neural network is estab⁃ lished
Summary
采用将三维坐标分别进行独立预测方法时,以 对 x 坐标进行预测为例,每组输入样本数据为 10×1 的列向量,每组输出样本数据为 1×1 的列向量。 此 时神经网络的输入节点数为 10,输出节点数为 1。 进行轨迹预测时,将三维坐标独立进行预测的相对 误差比将它们视为整体进行预测的相对误差小,验证了本文所提独立预测方法的可行性; HPSO⁃ TPFENN 取得了最好的预测效果,预测相对误差普 遍不超过 1%;Elman 网络取得了较好的预测效果, 但其误差明显高于 HPSO⁃TPFENN, 说明利用杂交 粒子群进行参数寻优并加入时间收益因子可以提高 网络的预测精度;而 BP 网络的预测误差相对较大, 说明传统的前馈神经网络不适合处理时间序列的预 测问题。 由图 11 可以看出,采用某一固定预测方法时, HPSO⁃TPFENN 的迭代次数比 TPFENN 和 Elman 的 迭代次数少;采用独立预测方法后,4 种模型的迭代 次数都明显下降。 这说明独立的预测方法、时间收 益因子的引入和 HPSO 的优化都有利于提高模型的 训练效率。 Track Prediction Based on Neural Networks and Genetic Algorithm[ J] .
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: Xibei Gongye Daxue Xuebao/Journal of Northwestern Polytechnical University
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.