Analysis of Multi-Sensor Fusion Technology Applied in Autonomous Navigation
Main Article Content
Keywords
multi-sensor fusion, autonomous navigation, SLAM, pose estimation, path planning, Kalman filtering
Abstract
Aiming at the problems of perception blind zones, error accumulation, and poor environmental adaptability inherent in single -sensor autonomous navigation systems, this paper investigates multi -sensor fusion technology for autonomous navigation. By integrating LiDAR, vision, IMU, and GNSS/RTK sensors, a unified framework encompassing spatio -temporal registration, feature extraction, data fusion, and global optimization is constructed. Extended Kalman Filter (EKF) is employed to achieve tightly -coupled fusion, while graph optimization is utilized for SLAM correction. Path planning is realized based on the Dynamic Window Approach (DWA) and the Timed Elastic Band (TEB) algorithm. Experiments conducted in multiple scenarios demonstrate that the system achieves a positioning accuracy of 0.08 m in normal environments, better than 0.12 m in complex scenes, a tunnel trajectory deviation error of 0.064 m, and a lateral tracking error of ± 2.6 cm for agricultural machinery. The success rates for static and dynamic obstacle avoidance are 100% and 95.7%, respectively. This method can significantly improve navigation accuracy and robustness in complex environments, providing technical support for the autonomous navigation of intelligent equipment.
References
- [1] Liang, L. (2026). Research on autonomous navigation technology of intelligent robots based on multi - sensor fusion. Intelligent IoT Technology, 58(1), 107–112.
- [2] Zhu, D. Y. (2026). Research on UAV autonomous navigation and obstacle avoidance technology in tunnels based on multi-sensor fusion. Technology Innovation and Application, (4), 27–30.
- [3] Ni, Y. Y. (2026). Design and testing of automatic driving control technology for agricultural tractors based on multi-sensor fusion. Farm Machinery Use and Maintenance, (1), 26–29.
- [4] Guo, S. H. (2025). Application of SLAM and AR multi -sensor fusion in autonomous driving technology. Automobile Era, (13), 13–15.
- [5] Xie, Z. X., Hu, Y. J., & Lin, H. Y. (2025). Research on intelligent fusion perception technology of automotive multi-sensors for autonomous driving. Popular Automobile, (11), 25–27.
- [6] Chen, D. D., Lu, Y., Huang, Z. J., et al. (2025). Research on automatic sensor fusion technology in autonomous driving based on the YOLO algorithm. Automobile Maintenance and Service, (9), 129.
- [7] Zhu, M. C., Lv, Z. J., Deng, M. H., et al. (2024). Autonomous navigation ROS intelligent robots based on multi-sensor fusion. Automation Application, 65(13), 1–4.
- [8] Li, Z. X. (2025). Application of multi -sensor information fusion and decision control in autonomous navigation of guide robots. Heilongjiang Science, 16(6), 19–22.
- [9] Wu, Z., & Yu an, P. T. (2023). Research on anti -deviation of tunnel belt conveyor systems based on multi-sensor fusion adaptive control. Yellow River, 45(S2), 144–145.
- [10] Sun, J. B., Liu, Z. J., Yang, F. Z., et al. (2023). Review of key technologies for agricultural equipment and sloping land operations in hilly and mountainous areas. Transactions of the Chinese Society for Agricultural Machinery, 54(5), 1–18.
- [11] Zhao, Z. G., Xiao, W., Mao, K. K., et al. (2024). Review of laser SLAM and path planning for inspection unmanned v ehicles. Journal of Mechanical Design and Manufacturing Engineering, 53 (8), 1–4.
- [12] Wang, W., Wang, C., Liu, J., et al. (2024). HVL-SLAM: Hybrid vision and LiDAR fusion for SLAM. IEEE Transactions on Geoscience and Remote Sensing.
- [13] Liu, Z., Fang, J., & Zhao, Y. (2025). DynaFusion-SLAM: Multisensor fusion and dynamic optimization of autonomous navigation algorithms for pasture-pushing robots. Sensors, 25(11), 3395.
- [14] Huang, Z., Ye, G., Yang, P., et al. (2025). Application of multisensor fusion localization algorit hms based on recurrent neural networks. Scientific Reports, 15(1), 1142–1151.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.