GAF-PCNN: A Multisensor Fusion Navigation Integrity Monitoring Method Based on Gramian Angular Field and Parallel CNN

To address the issue of navigation integrity loss caused by navigation sensor failures during the actual operation, a multisensor fusion navigation integrity monitoring method based on a Gramian angular field (GAF)-parallel convolutional neural network (PCNN) is proposed. First, according to the subsystem combination in the multisensor fusion navigation system, principal component analysis (PCA) is used to extract the fault principal component features from the least-squares form of Kalman filter (KF-LS) form solution and establish fault detection statistics, forming 1-D time-series feature data. Second, the GAF is used to convert the 1-D time-series feature data, generating two sets of image data containing state features and time information. These two sets of image data are then fed into a dual-channel convolutional neural network (CNN) for parallel training. In a real-time system, the trained model is used for fault detection and isolation. Finally, the system protection level (PL) is calculated, and the integrity assessment and warning of the multisensor fusion navigation system are realized. In the marine environment, simulation tests were conducted using real collected ship trajectory data. The test results show that the method can promptly and effectively detect and isolate faults. The fault classification accuracy on the test set is 96.9%, the detection time for constant bias faults is around 1 s, and the detection time for gradual faults is around 3.5 s. The calculated PL can issue integrity warning prompts within 5 s when the positioning error exceeds the alarm threshold of 5 m, ensuring the robustness and reliability of the multisensor fusion navigation system.