LSTM-based track prediction method

• Main Authors: HAN XU, WANG XUDONG, WANG YUHANG, GUO ZONGHAO, SUN QITAI, WANG GUOQI
• Format: Patent
• Language: Chinese; English
• Published: 17.10.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; GYROSCOPIC INSTRUMENTS; MEASURING; MEASURING DISTANCES, LEVELS OR BEARINGS; NAVIGATION; PHOTOGRAMMETRY OR VIDEOGRAMMETRY; PHYSICS; SURVEYING; TESTING

The invention discloses a track prediction method based on LSTM (Long Short Term Memory). The track prediction method comprises the following steps: step 1, obtaining and preprocessing a track sequence data set; wherein the acquisition and preprocessing of the track sequence data set aims to process acquired ADS-B original data into a required stable multivariable time sequence data set with uniform length; step 2, constructing an LSTM track prediction model, wherein the construction of the LSTM track prediction model aims at performing training learning on the acquired ADS-B data through an LSTM model to realize track prediction; 3, inputting a track to be detected into the model to obtain a predicted track; inputting a to-be-detected track into the model to obtain a predicted track so as to realize a track prediction function by using the model. According to the LSTM track prediction model combined with the ADS-B data, the difference between the prediction result and the actual value is small, the fitting d