Scholarlink

<p>When Global Positioning System (GPS) signals are unavailable due to outages or interference, the GPS/Inertial Navigation System (INS) integrated navigation system relies solely on the INS, leading to error accumulation and degraded navigation accuracy. To address this challenge, a novel prediction model based on a Shallow Transformer (S-Transformer) is proposed. The model employs a shallow architecture combined with a Multi-Head Self-Attention Mechanism to effectively capture long-range dependencies, enabling real-time and accurate predictions of GPS pseudo-position and velocity increments during GPS signal outages, thereby correcting INS errors. Furthermore, to improve filtering accuracy under large prediction errors in the GPS pseudo-position and velocity, an Adaptive Square Root Cubature Kalman Filter (A-SRCKF) algorithm is developed. Leveraging the high short-term accuracy of INS, the algorithm adaptively estimates the GPS noise covariance, significantly reducing noise impact and enhancing GPS/INS navigation precision. Experimental results from both simulation and real-world road scenarios validate that integrating the S-Transformer model with the A-SRCKF algorithm substantially improves navigation system accuracy and robustness, achieving superior performance over other state-of-the-art approaches. The development of integrated navigation systems capable of maintaining high precision and reliability in GPS-denied environments is fundamentally supported by this study.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel real-time integrated navigation system based on the S-Transformer and A-SRCKF during GPS outages

  • 期刊论文

  • 发表日期：

    2025年6月14日

查看全文
  • Kaifei He,
  • Shengwei Zhang,
  • Chenguang Yao,
  • Yue Wang,
  • Kai Ding

摘要

When Global Positioning System (GPS) signals are unavailable due to outages or interference, the GPS/Inertial Navigation System (INS) integrated navigation system relies solely on the INS, leading to error accumulation and degraded navigation accuracy. To address this challenge, a novel prediction model based on a Shallow Transformer (S-Transformer) is proposed. The model employs a shallow architecture combined with a Multi-Head Self-Attention Mechanism to effectively capture long-range dependencies, enabling real-time and accurate predictions of GPS pseudo-position and velocity increments during GPS signal outages, thereby correcting INS errors. Furthermore, to improve filtering accuracy under large prediction errors in the GPS pseudo-position and velocity, an Adaptive Square Root Cubature Kalman Filter (A-SRCKF) algorithm is developed. Leveraging the high short-term accuracy of INS, the algorithm adaptively estimates the GPS noise covariance, significantly reducing noise impact and enhancing GPS/INS navigation precision. Experimental results from both simulation and real-world road scenarios validate that integrating the S-Transformer model with the A-SRCKF algorithm substantially improves navigation system accuracy and robustness, achieving superior performance over other state-of-the-art approaches. The development of integrated navigation systems capable of maintaining high precision and reliability in GPS-denied environments is fundamentally supported by this study.