Scholarlink

This paper addresses the challenge of integrating the Global Positioning System (GPS) and Inertial Navigation System (INS) for navigating intelligent systems, particularly in urban environments where GPS outages are frequent. The INS is based on the dead reckoning principle to provide estimates of position, acceleration, attitude, and velocity without external input. However, its accuracy degrades over time due to cumulative errors. On the other hand, GPS offers high accuracy by utilizing satellite signals but is susceptible to blockages in environments such as tunnels or urban canyons. To mitigate the limitations of both systems, we propose an integration approach using the Extended Kalman Filter (EKF). Simulation results demonstrate significant improvements in navigation accuracy. The proposed EKF-based method reduces Root Mean Square Error (RMSE) in the East direction from 2.92 m (GPS) and 11.72 m (INS) to 1.34 m. In the North direction, RMSE decreases from 3.99 m (GPS) and 138.70 m (INS) to 1.87 m, while in the Up direction, it is reduced from 7.39 m (GPS) and 190.31 m (INS) to 3.56 m. Overall, our approach improves accuracy by 93.11% compared to standalone INS and 53.02% compared to standalone GPS.

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

Secure Navigation for Intelligent Systems in Urban Environments: GPS Data and INS Raw Measurements Integration Using EKF to Mitigate GPS Failure

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Mohammed Aftatah,
  • Khalid Zebbara,
  • Soufiane El Asri,
  • Asmaa Benchama

摘要

This paper addresses the challenge of integrating the Global Positioning System (GPS) and Inertial Navigation System (INS) for navigating intelligent systems, particularly in urban environments where GPS outages are frequent. The INS is based on the dead reckoning principle to provide estimates of position, acceleration, attitude, and velocity without external input. However, its accuracy degrades over time due to cumulative errors. On the other hand, GPS offers high accuracy by utilizing satellite signals but is susceptible to blockages in environments such as tunnels or urban canyons. To mitigate the limitations of both systems, we propose an integration approach using the Extended Kalman Filter (EKF). Simulation results demonstrate significant improvements in navigation accuracy. The proposed EKF-based method reduces Root Mean Square Error (RMSE) in the East direction from 2.92 m (GPS) and 11.72 m (INS) to 1.34 m. In the North direction, RMSE decreases from 3.99 m (GPS) and 138.70 m (INS) to 1.87 m, while in the Up direction, it is reduced from 7.39 m (GPS) and 190.31 m (INS) to 3.56 m. Overall, our approach improves accuracy by 93.11% compared to standalone INS and 53.02% compared to standalone GPS.