To improve the accuracy of personnel positioning in underground coal mines, in this paper, we propose a convolutional neural network (CNN) three-dimensional (3D) visible light positioning (VLP) system based on the Inception-v2 module and efficient channel attention mechanism. The system consists of two LEDs and four photodetectors (PDs), with the four PDs on the miner’s helmet. Considering the height fluctuation of PD and the impact of wall reflection on the received light power, we adopt the Inception module to perform a multi-scale extraction of the features of the received light power, thus solving the limitation of the single-scale convolution kernel on the positioning accuracy. In order to focus on the information that is more critical to positioning among the numerous input features, giving different features of the optical power data corresponding weights, we use an efficient channel attention mechanism to make the positioning model more accurate. The simulation results show that the average positioning error of the system was 1.63 cm in the space of 6 m × 3 m × 3.6 m when both the line-of-sight (LOS) and non-line-of-sight (NLOS) links were considered, with 90% of the localization errors within 4.55 cm. During the experimental stage, the average positioning error was 11.12 cm, with 90% of the positioning errors within 28.75 cm. These show that the system could achieve centimeter-level positioning accuracy and meet the requirements for underground personnel positioning in coal mines.
2 September 2025 in Measurement Science and Technology
In recent years, visible light-based indoor positioning technology has become a prominent research topic within the academic community. To address the issues of high localization errors in low-light and corner regions of indoor environments, this paper proposes a solution based on ordering points to identify the clustering structure (OPTICS) clustering coupled with partitioned linear interpolation. The proposed approach first segments the indoor area into zones; within dark regions, the OPTICS clustering algorithm is employed to identify zones of dense optical power, followed by the correction of discrete point light intensity data to reduce localization errors in low-light areas. In corner regions, edge data points are used for linear interpolation to enhance the feature set and improve positioning accuracy. To enhance neural network training efficiency, a hybrid method combining an adaptive JAYA algorithm (AJAYA) with update gate recurrent neural network (UGRNN) is introduced. This approach leverages AJAYA for efficient hyperparameter tuning of the UGRNN, thereby improving training speed and system robustness. To overcome the challenge of differentiating between unmodulated LED devices, radio frequency identification (RFID) technology is integrated with the visible light positioning system by deploying ultra-high frequency RFID tags to enable effective distinction of different LEDs. Simulation results demonstrate that, within a large space of 7 m × 4 m × 4.6 m, the system achieves an average localization error of 0.08 m. Experimental validation shows that the static positioning error averages 0.28 m, and dynamic positioning accuracy is improved by 35.14% after using inertial measurement unit data to correct the light intensity measurements. Additionally, applying rectangular trajectory tracking yields a 30.2% improvement in positioning accuracy. These results verify the effectiveness and stability of the proposed method in complex indoor scenarios.