LiDAR Sensors Face Challenges in Rainy and Snowy Weather Conditions

Strategic Insights -

Researchers from the Nanjing University of Aeronautics and Astronautics have identified a crucial limitation of LiDAR sensors in high-level self-driving cars. The sensors struggle to accurately detect and track objects in wet weather conditions due to the absorption and diffraction of rain and snow particles, which weakens the point clouds. To address this issue, the researchers proposed a multi-object detection and tracking strategy that leverages regional feature enhancement and a multicategory tracking module integrated with an unscented Kalman filter. The strategy demonstrates improved accuracy and stability in object tracking under weak point cloud features.

Key Takeaways:

  • LiDAR sensors are impaired in rainy and snowy weather conditions, reducing their accuracy and reliability in object detection and tracking.
  • The proposed multi-object detection and tracking strategy addresses the shortcomings of LiDAR sensors by enhancing regional features and integrating a multicategory tracking module.
  • Simulation results show that the proposed method achieves an AMOTA of 74.40%, surpassing mainstream methods such as Poly-MOT and Fast-Poly.
  • The research was financially supported by the National Natural Science Foundation of China (NSFC), the Youth Foundation of Jiangsu Province, and the Jiangsu Funding Program for Excellent Postdoctoral Talent.
  • The proposed strategy demonstrates improved performance in noisy and weak point cloud environments, making it a promising solution for high-level self-driving cars.

Statistics:

  • AMOTA (Accuracy in Multi-Object Tracking and Analysis) of the proposed method: 74.40%
  • Comparison to mainstream methods: Poly-MOT: 68.50%, Fast-Poly: 72.10%

Sources:

  • "Vehicle-mounted Lidar Multiobject Detection and Tracking Under Weak Point Cloud" (Ieee Sensors Journal, 2025;25(16):31611-31623)
  • Wanzhong Zhao et al.'s research, funded by the National Natural Science Foundation of China (NSFC), the Youth Foundation of Jiangsu Province, and the Jiangsu Funding Program for Excellent Postdoctoral Talent.