This post is also available in:
עברית (Hebrew)
GNSS (Global Navigation Satellite Systems) plays a pivotal role in modern technologies like autonomous vehicles and intelligent transportation systems. However, in urban areas, tall buildings, vehicles, and other structures often obstruct satellite signals, leading to Non-Line-of-Sight (NLOS) errors. These errors can significantly impact positioning accuracy, which is crucial for smart city infrastructure and navigation technologies. Now, researchers have introduced an innovative artificial intelligence (AI)-based solution to detect and mitigate these errors, offering a promising advance in urban navigation.
The study, Published in Satellite Navigation, presents a novel machine-learning approach designed to tackle NLOS issues in GNSS systems. Developed by researchers from Wuhan University, Southeast University, and Baidu, the method uses the AI model Light Gradient Boosting Machine (LightGBM), the model can differentiate between Line-of-Sight (LOS) and NLOS signals, improving the reliability of GNSS-based positioning systems.
According to TechXplore, The researchers validated their solution through real-world experiments conducted in Wuhan, China, testing it in complex urban environments with dense buildings. The method employs a fisheye camera to capture satellite visibility, categorizing signals as either LOS or NLOS. It then analyzes features such as signal-to-noise ratio, elevation angle, pseudorange consistency, and phase consistency. By identifying correlations between these factors and the signal types, the LightGBM model achieved a 92% accuracy, outperforming traditional methods.
Dr. Xiaohong Zhang, the lead researcher, highlighted the potential impact of this innovation: “By using machine learning to analyze multiple signal features, we’ve shown that excluding NLOS signals can significantly boost the accuracy and reliability of satellite-based navigation systems. This has profound implications for applications such as autonomous driving and smart city infrastructure.”
This breakthrough promises to enhance navigation in cities, ensuring that GNSS systems remain reliable even in challenging environments. As urban areas become smarter and more connected, improving the precision of GNSS technology will be essential for the next generation of transportation and urban infrastructure.
