A research team from the Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, proposed a real-time smoothing algorithm to improve the accuracy and stability of photoelectric theodolites. Published in Scientific Reports by Nature Portfolio, the study addressed challenges in outlier detection and data interpolation for external guidance systems. Photoelectric theodolites are widely used in range experiments due to their high precision and anti-interference capabilities. However, external guidance data often contains outliers caused by environmental noise or equipment limitations, and low data frequency can lead to discontinuous tracking. To solve these issues, the team introduced a dynamic threshold-based outlier elimination method and an adaptive interpolation algorithm.
For outlier detection, the researchers employed an influence function to calculate sample variance dynamically, allowing the system to distinguish normal data from outliers more accurately. Experiments showed an average detection rate of over 80% with a low false alarm rate. For interpolation, the algorithm categorized "stuck" data—where consecutive frames had identical values—and applied different smoothing strategies. This ensured smooth transitions between data points, preventing abrupt movements that could damage the servo system.
The algorithm was tested in both simulations and real-world naval range experiments. Compared to fixed-threshold and non-impact function methods, the dynamic threshold approach improved outlier rejection by up to 127.92% while reducing false alarms. The interpolation method also minimized angular velocity fluctuations, making it safer for servo systems.
This study provides a practical solution for real-time data processing in photoelectric theodolites, with potential applications in radar and satellite trajectory measurement. Future work may extend the algorithm to other data sources, further enhancing tracking stability.
中文
