In order to achieve the precise positioning of railway inspection robots and optimize their inspection paths, this paper focuses on robot work optimization for railway inspection environments. Firstly, an adaptive improvement strategy is proposed for the extended Kalman filter algorithm, and a multi-sensor information fusion-based positioning algorithm is implemented for railway inspection robots. Then, based on the improved ant colony optimization algorithm, a path planning model for inspection robots is constructed. The experimental results show that the multi-sensor information fusion-based positioning algorithm features a high positioning accuracy, and the positioning root mean square error converges to the minimum value of 0.136. To that, the inspection robot localization success rate and the stability of the proposed method make it better than other methods, with an increase in the success rate amounting to 29.41%. At the same time, the improved ant colony optimization-based path planning model achieves the highest planning efficiency, with a localization success rate of 97.35% in railway inspection environments with obstacles, a maximum reduction of the inspection path length amounting to 36.49%, an average planning time reduction of up to 46.67 seconds, and excellent path smoothness. The implementation of this positioning algorithm and path planning model helps improve the railway inspection efficiency and promotes the development of multi-sensor fusion technology.
Inspection robot, Railway, Information fusion, Sensor, Ant colony optimization algorithm.
YunFei WANG, MingLiang LIANG, YaRu ZHAO, "Localization and Path Planning for Railway Inspection Robot Based on Multi-Sensor Fusion and Improved ACO", Studies in Informatics and Control, ISSN 1220-1766, vol. 34(4), pp. 29-40, 2025. https://doi.org/10.24846/v34i4y202503
All articles published by Studies in Informatics and Control (SIC) are published in accordance with the terms and conditions of the Creative Commons Attribution License (CC BY-NC-SA 4.0).