GNSS-INS Fusion for Enhanced Positioning and Signal Processing in Railway Track Inspection Under Dynamic Conditions

Abstract

The data collection and processing technology of orbit detectors and global navigation satellite systems in driving conditions is of great significance. A positioning method based on the fusion of inertial navigation system and global navigation satellite system is proposed to address the signal acquisition and fusion issues between the track inspection instrument and the global navigation satellite system during driving. This method achieves effective fusion of multi-source data through synchronous acquisition technology and time alignment algorithm, and further optimizes the data processing flow of the track inspection instrument, enhancing positioning accuracy and robustness. The experiments tested the performance of different methods in terms of positioning error, redundancy detection rate, sampling accuracy, and resource consumption by simulating three typical scenarios: open environment, urban rail transit, and mountain track. The results showed that the proposed method had an average positioning error of only 2.5 mm in an open environment, which was significantly better than the 6.2 mm error of GNSS-RTK and the 12.5 mm error of GNSS. The redundancy detection rate could reach 95.2%, which was nearly 10% higher than GNSS-RTK and 30% higher than GNSS. In urban and mountainous environments, positioning errors are kept within 6.8 and 8.5 mm, respectively. Meanwhile, the research's proposed method has improved its signal-to-noise ratio to 45.8 dB and decreased its mean square error by over 40%. This demonstrates its excellent anti-interference and denoising capabilities. The experimental results demonstrate that the proposed method significantly improves the positioning accuracy and real-time performance of the orbit detection system under complex operating conditions. The method is applicable to engineering and has significant promotional value.