Flight trajectory anomaly detection method based on LSTM-GBSVDD model
By using the LSTM-GBSVDD model, which combines the LSTM network and the SVDD algorithm, the problems of varying flight trajectory data length and complexity are solved, achieving efficient and accurate flight trajectory anomaly detection. This improves detection accuracy and efficiency and is suitable for unsupervised anomaly detection of complex time-series data.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NAVAL UNIV OF ENG PLA
- Filing Date
- 2024-12-02
- Publication Date
- 2026-05-01
AI Technical Summary
Existing flight trajectory anomaly detection methods suffer from inaccurate feature extraction and low detection efficiency when dealing with complex, dynamic flight trajectory data with real-time changes in data length. They also struggle to effectively capture long-term dependencies in the trajectory, resulting in insufficient detection accuracy.
An anomaly detection method based on the LSTM-GBSVDD model is adopted. The time series features are extracted by the LSTM network and the feature mean pooling is performed. A multidimensional hypersphere classifier is constructed by combining the SVDD model. Gradient optimization is used to train the LSTM and SVDD parameters to achieve joint optimization and improve detection accuracy and efficiency.
It effectively handles variable-length data sequences, improves the model's universality and detection accuracy, significantly enhances the accuracy and computational efficiency of anomaly detection, can efficiently identify potential abnormal trajectories, reduces false alarm rates, and is suitable for anomaly detection of flight trajectories in unsupervised environments.
Smart Images

Figure CN119848536B_ABST
Abstract
Citation Information
Patent Citations
Unmanned aerial vehicle flight data abnormity detecting method based on LSTM
CN108960303A
Water treatment time sequence data anomaly detection method based on LSTM
CN112765896A