Optimization Research on Autonomous Driving Environmental Perception Algorithm for Complex Weather (Heavy Rain/Heavy Snow/Fog)

Xijie Hong

doi:10.47297/taposatWSP2633-456912.20260704

PDF
导出
分享
收藏
专辑

Optimization Research on Autonomous Driving Environmental Perception Algorithm for Complex Weather (Heavy Rain/Heavy Snow/Fog)

7卷第4期

浏览量: 0 Downloads: 0

Optimization Research on Autonomous Driving Environmental Perception Algorithm for Complex Weather (Heavy Rain/Heavy Snow/Fog)
Optimization Research on Autonomous Driving Environmental Perception Algorithm for Complex Weather (Heavy Rain/Heavy Snow/Fog)
School of Transportation, Fujian University of Technology, Fuzhou, Fujian, 350118, China
2026年7卷第4期页码：71-77
纸质出版：2026-04-30
DOI：10.47297/taposatWSP2633-456912.20260704
Accepted：

Scan QR Code

Xijie Hong. (2026). Optimization Research on Autonomous Driving Environmental Perception Algorithm for Complex Weather (Heavy Rain/Heavy Snow/Fog). 科技理论与实践（英文版）, 7(4), 71-77.

Hong Xijie.Optimization Research on Autonomous Driving Environmental Perception Algorithm for Complex Weather (Heavy Rain/Heavy Snow/Fog)[J].Theory and Practice of Science and Technology,2026,07(04):71-77.
Xijie Hong. (2026). Optimization Research on Autonomous Driving Environmental Perception Algorithm for Complex Weather (Heavy Rain/Heavy Snow/Fog). 科技理论与实践（英文版）, 7(4), 71-77. DOI： 10.47297/taposatWSP2633-456912.20260704.

Hong Xijie.Optimization Research on Autonomous Driving Environmental Perception Algorithm for Complex Weather (Heavy Rain/Heavy Snow/Fog)[J].Theory and Practice of Science and Technology,2026,07(04):71-77. DOI： 10.47297/taposatWSP2633-456912.20260704.

摘要

Abstract

With the increasing number of traffic accidents caused by complex weather (heavy rain/snow/fog)

complex weather also leads to the core problem of performance degradation in autonomous driving environmental perception. To improve the autonomous driving environmental perception algorithm

a three-level optimization framework of "weather interference quantification model - single sensor anti-interference preprocessing - multi-modal dynamic fusion" is constructed to enhance the robustness of the perception algorithm. The study adopts a combination of data-driven and model innovation methods

and is expected to increase the target detection accuracy under severe weather by more than 25%

providing technical support for the all-weather safe operation of L4-level autonomous driving

and having both theoretical breakthroughs and industrial application value.

关键词

Keywords

references

Liu, W., et al. (2021). SSD: Single Shot MultiBox Detector. InEuropean Conference on Computer Vision. Springer.

He, K., et al. (2020). Deep Residual Learning for ImageRecognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(3), 789-801.

Chen, X., et al. (2022). Robust Multimodal Fusion for AutonomousDriving in Adverse Weather. IEEE Transactions on Intelligent Vehicles, 7(4), 1123-1135.

Chen, L., Wang, Q., Li, N. (2023). Complex weather multi-modalfusion perception algorithm for autonomous driving. Automotive Engineering, 45 (6): 987-995.

Zhang, R., et al. (2023). Attention-Based Fusion of LiDAR andCamera for All-Weather Perception. IEEE Sensors Journal, 23(8), 8456 - 8468.

Zhang, W., Liu, M. (2022). LiDAR and image feature fusion methodbased on attention mechanism. Chinese Journal of Computers, 45 (8): 1765-1778.

Li, Y., Wang, H., Zhang, C. (2021). Robust environmentalperception for autonomous driving in heavy rain. IEEE Transactions on Intelligent Transportation Systems, 22 (5): 2890 - 2899.

Wang, Y., et al. (2023). Adverse Weather Dataset for AutonomousDriving: Rain, Snow, and Fog. IEEE International Conference on Robotics and Automation (ICRA).

Li, J., et al. (2024). Infrared Thermal Imaging for AutonomousDriving in Fog and Rain. IEEE Transactions on Vehicular Technology, 73(2), 1125-1137.

Zhou, Y., Chen, X., Lin, T. (2025). SPG: Semantic Point CloudGeneration for Robust 3D Object Detection in Adverse Weather. IEEE Transactions on Intelligent Transportation Systems, 26(9): 4210 - 4223.