Global Navigation Satellite Systems (GNSS), such as Global Positioning System (GPS) , are fundamental for Unmanned Aerial Vehicle (UAV) navigation, but their effectiveness diminishes in urban canyons, indoor spaces, or hostile environments where signals are obstructed or jammed. Traditional reliance on GPS leaves drones vulnerable to disruptions, whether from tall buildings, weather conditions, or cyberattacks. While alternatives like inertial sensors and LiDAR show promise, they often struggle with drift and computational demands. Vision-based and terrain-aided systems could offer viable solutions, but they need further refinement to adapt to dynamic environments. This highlights the pressing need for robust multi-sensor fusion frameworks to enable autonomous and safe UAV operations in GPS-denied areas.
Published (DOI: 10.1186/s43020-025-00162-z) on April 7, 2025, in Satellite Navigation, a research team from Prince Sultan University provides a comprehensive review of UAV navigation in GPS-denied environments. The review evaluates 132 papers, focusing on absolute and relative localization techniques, including vision-based systems, LiDAR, and terrain-aided algorithms. By examining computational efficiency and sensor fusion, the research identifies hybrid approaches as the most reliable solution for UAV navigation. This work bridges significant gaps in existing technologies, offering practical insights for real-world applications where GPS signals are unavailable.
The review examines two primary methods for UAV navigation in GPS-denied areas: absolute localization, which uses pre-mapped terrain data (e.g., TERCOM and DSMAC), and relative localization methods like SLAM (Simultaneous Localization and Mapping) and visual-inertial odometry that rely on real-time sensor data. While absolute methods face limitations in featureless environments, relative techniques offer adaptability but require significant computational resources. Vision-based systems, particularly when enhanced with AI for feature recognition, hold considerable promise, though lighting conditions remain a challenge. The research emphasizes the importance of sensor fusion, demonstrating that combining LiDAR, radar, and inertial measurements, alongside advanced filtering techniques such as Kalman filters, can substantially improve navigation reliability. Furthermore, real-time processing is crucial, with hardware accelerators like GPUs and optimized algorithms (such as LSTM networks) enabling faster data analysis and decision-making. While hybrid systems combining terrain maps with live SLAM data offer a balance of accuracy and flexibility, the study acknowledges the need for further refinement to scale these solutions across various environments. Advancements in AI processing power and edge computing will be key to fully autonomous UAV operations in unpredictable real-world conditions.
Dr. Imen Jarraya, lead author of the study, emphasized, "No single sensor or algorithm can solve all the challenges of GPS-denied navigation. Our research shows that combining absolute and relative localization with multi-sensor fusion is the key to achieving reliable UAV navigation. Future work must focus on optimizing these systems to handle the unpredictability of environments ranging from dense urban areas to remote disaster zones."
This research holds significant implications for industries relying on UAVs, such as logistics, agriculture, and defense. UAVs delivering medical supplies to remote or disaster-stricken areas could operate without GPS, and military drones could navigate in signal-jammed regions. The study also points to the need for regulatory frameworks to standardize these technologies, ensuring their safe and efficient integration into future infrastructures. As UAVs become integral to smart cities and infrastructure inspection, overcoming the limitations of GPS will ensure safer, more effective operations. These findings encourage further investment in AI-driven navigation and collaborative research to refine these systems for global use.
###
References
DOI
10.1186/s43020-025-00162-z
Original Source URL
https://doi.org/10.1186/s43020-025-00162-z
Funding information
The research was funded by PSDSARC seed project number (PSDSARC Project ID: PID-000085_01_02), and the APC was funded by PSU.
About Satellite Navigation
Satellite Navigation (E-ISSN: 2662-1363; ISSN: 2662-9291) is the official journal of Aerospace Information Research Institute, Chinese Academy of Sciences. The journal aims to report innovative ideas, new results or progress on the theoretical techniques and applications of satellite navigation. The journal welcomes original articles, reviews and commentaries.