Aircraft maneuvering trajectory prediction method based on Morifier-BiGRU

Abstract

The invention provides an aircraft maneuvering trajectory prediction method based on the Morifier-BiGRU, the BiGRU and the aircraft maneuvering trajectory prediction problem are combined, the problemof coupling loss between output data and input data of a hidden layer of a loop network is fully considered, a Morifier coupling function is introduced, and the problem that historical information isdiscarded by mistake in the weight parameter updating process is solved. According to the method, aircraft maneuvering trajectory data prediction can be well achieved, the accuracy and advancement ofsix-degree-of-freedom maneuvering trajectory data prediction are greatly improved, meanwhile, the network model has no limitation on the sequence length and dimension of incoming data, has high adaptability to the data, and reliable support is provided for judging the position and attitude information of the target aircraft in advance.

Description

technical field [0001] The invention relates to the fields of aircraft flight control and artificial intelligence, in particular to an aircraft multi-dimensional maneuver trajectory prediction method. Background technique [0002] With the rapid development of aviation technology, the prediction technology of maneuvering and trajectory data of the target aircraft has become one of the hot spots in the field of scientific research all over the world. Traditional aircraft maneuver trajectory data prediction often adopts linear regression method, differential autoregressive moving average model, hidden Markov process, time series decomposition method, etc., but these methods cannot fully analyze the correlation between various data dimensions, so in the face of When using high-dimensional maneuver trajectory data, its prediction accuracy has certain limitations. In recent years, due to the rapid development of deep learning methods and artificial intelligence theory, it has sh...

AI Technical Summary

Problems solved by technology

At the same time, the present invention introduces the Mogrifier coupling module in the forward and backward data propagation process of the network to solve the problem that the historical information is wrongly discarded in the weight parameter update process

Images