Preliminary Study on Neural Network-Based Insulator Detection for an Insulator-Cleaning Robot via MQTT with CPU-Based Processing
Article Sidebar
Main Article Content
Abstract
Object detection is a process that utilizes techniques from neural networks, which have gained significant popularity in robot vision research. This study applies such techniques to an insulator cleaning robot, where images from a camera are processed to determine the target’s bounding box coordinates. The coordinates are then transmitted to the robot’s control unit via long-range wireless communication using MQTT.The experiments were conducted on a 6-Core Intel Xeon E5 CPU (3.5 GHz) with 16 GB of RAM, without using a GPU for model inference. Among several tested architectures, YOLOv11n delivered the best performance, achieving an mAP@0.5 of 97.6% and a precision of 98.66%. The overall system speed (including both inference and data transmission) averaged 1.61 FPS. While this is slower than previous research reporting 24–51 FPS, the developed system offers a unique advantage: it can transmit target coordinates 0.62 seconds in advance, allowing the control unit to navigate toward the target without relying on real-time image processing at every step. Furthermore, the combined processing and transmission time accounted for only 52% of the total per-frame time, demonstrating the system’s suitability for practical applications—even when running solely on a CPU.
Downloads
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Journal of TCI is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) licence, unless otherwise stated. Please read our Policies page for more information...
References
J. Arents and M. Greitans, "Smart industrial robot control trends, challenges and opportunities within manufacturing," Applied Sciences, vol. 12, no. 2, p. 937, 2022.
M. Casini, D. Prattichizzo, and A. Vicino, "The automatic control telelab," IEEE Control Systems Magazine, vol. 24, no. 3, pp. 36-44, 2004.
J. Qi et al., "Computer vision-based hand gesture recognition for human-robot interaction: a review," Complex & Intelligent Systems, vol. 10, no. 1, pp. 1581-1606, 2024.
Z. Zou et al., "Object detection in 20 years: A survey," Proceedings of the IEEE, vol. 111, no. 3, pp. 257-276, 2023.
C. Ju et al., "Speed-FairMOT: multi-class multi-object tracking for real-time manipulation," Signal, Image and Video Processing, 2025.
Y. Ye et al., "Dynamic and real-time object detection based on deep learning for home service robots," Sensors, vol. 23, no. 23, p. 9482, 2023.
L. He et al., "Research and application of YOLOv11-based object segmentation," Buildings, vol. 14, no. 12, p. 3777, 2024.
S. Zhai et al., "DF-SSD: An improved SSD object detection algorithm," IEEE Access, vol. 8, pp. 24344-24357, 2020.
Y. Wang et al., "Automatic ship detection based on RetinaNet," Remote Sensing, vol. 11, no. 5, p. 531, 2019.
P. Jiang et al., "A review of YOLO algorithm developments," Procedia Computer Science, vol. 199, pp. 1066-1073, 2022.
M. Sportelli et al., "Evaluation of YOLO object detectors for weed detection," Applied Sciences, vol. 13, no. 14, p. 8502, 2023.
W. Luangluewut et al., "Performance Comparison of LRT and YOLO," Defence Technology Academic Journal, 2025.
C. Kang and S. Kim, "Real-time object detection and segmentation technology," JMST Advances, vol. 5, no. 2, pp. 69-76, 2023.
T. Diwan et al., "Object detection using YOLO," Multimedia Tools and Applications, vol. 82, no. 6, pp. 9243-9275, 2023.
D. Maji et al., "YOLO-pose," CVPR, pp. 2637-2646, 2022.
W. Luangluewut et al., "Domain Adaptation for Drone Detection," ICAIIC, 2025.
R. Khanam and M. Hussain, "YOLOv11: An overview," arXiv, 2024.
G. Jocher et al., "Ultralytics YOLO11," GitHub, 2024.
J. Redmon et al., "You Only Look Once," CVPR, pp. 779-788, 2016.
M. Tan et al., "EfficientDet," CVPR, pp. 10781-10790, 2020.
B. Mehlig, Machine Learning with Neural Networks. Cambridge University Press, 2021.
OASIS, "MQTT Version 3.1.1," 2014.
S. Visa et al., "Confusion Matrix-based Feature Selection," CEUR Workshop Proc., 2011.
Roboflow, "Insulator Detection Dataset," 2023.
