Integrated INS/GNSS Navigation Systems: A Comprehensive Review of Filtering and AI-Based Fusion Techniques

Document Type : Review Article

Authors

1 Electrical Engineering Department, Faculty of Engineering, Suez Canal University, Ismailia, Egypt

2 College of Computing and IT, Arab Academy for Science, Technology& Maritime Transport, Cairo, Egypt

Abstract

Integrated Inertial Navigation System (INS) and Global Navigation Satellite System (GNSS) architectures have become essential for modern autonomous and navigation applications, offering complementary strengths to address the limitations of standalone systems. However, the fusion of INS and GNSS data presents several challenges, including handling sensor drift, nonlinearity, GNSS signal outages, and system uncertainties. This review systematically explores the current state of INS/GNSS integration, emphasizing the classification of fusion architectures (loosely, tightly, and ultra-tightly coupled) and the diverse inertial sensor technologies employed, including Micro-Electromechanical Systems (MEMS), fiber optic gyroscopes (FOG), and ring laser gyroscopes (RLG). Special attention is given to data fusion techniques, highlighting both classical model-based filters (e.g., Kalman Filter and its variants) and emerging artificial intelligence (AI)-based methods such as deep learning and recurrent neural networks. The paper also examines AI’s role in replacing or augmenting traditional filters and the use of platform-specific motion constraints to improve localization accuracy. This review aims to guide researchers and engineers in designing robust, intelligent navigation systems suited for dynamic and GNSS-challenged environments by synthesising advancements in filtering algorithms, AI techniques, and sensor technologies.

Keywords

Main Subjects

Files

Share

How to cite

Statistics