From Nominal Intensity to Equivalent Rainfall: A Path-Based Credibility Evaluation Framework for Simulated Rainfall in Autonomous-Driving Perception Tests

2026-06-10

AI Summary

This study addresses the limitations of current autonomous driving perception testing, where simulated rainfall often relies on nominal intensity or single-point measurements that fail to accurately capture real-world rain effects on perception systems. To bridge this gap, the authors propose a path-level fidelity assessment method that introduces, for the first time, path-equivalent rainfall intensity and a Raindrop Distribution Realism (RRD) score, integrated with LiDAR point cloud count and average reflectivity for perception consistency calibration. Grounded in joint modeling of real raindrop size distributions, spatial path sampling, and uncertainty quantification, the approach establishes an interpretable mapping from nominal parameters to actual perceptual impacts. Experimental evaluation within a 2.4 m × 7.2 m test area identifies Path IV (11.54 ± 0.31 mm/h, RRD=0.43) and Path VI (8.28 ± 0.34 mm/h, RRD=0.46) as optimal trajectories, demonstrating balanced performance in rainfall intensity stability, drop size distribution fidelity, and perception consistency.

Abstract

Credible simulated-rainfall conditions are essential for identifying perception-system boundaries and supporting SOTIF-oriented risk assessment in automated driving. However, closed-field tests are often described only by nominal rainfall intensity or single-point measurements, making it difficult to align simulated rain fields with real rainfall and map test results to real-world scenarios. This paper proposes a path-based credibility evaluation method for simulated rainfall in autonomous-driving perception tests. Using the drop size and velocity joint distribution of real rainfall as the reference, each candidate path is represented by path-equivalent rainfall intensity, an uncertainty band, and a path-averaged Realism of Raindrop Distribution (RRD) score. Lidar target point-cloud count and mean reflectivity are further used for perception-consistency correction, quantifying the proxy capability of each simulated-rainfall path for real-rainfall perception effects. Experiments are conducted using about 10,000 real-rainfall raindrop-spectrum samples, 728 RainSense perception samples, and 45 spatial sampling points in a 2.4 m x 7.2 m simulated-rainfall area. Results show that spatial non-uniformity remains under the same nominal condition, confirming the need for path-based evaluation. The method identifies Path IV and Path VI as preferable candidates, with results of 11.54 +/- 0.31 mm/h, RRD = 0.43, and 8.28 +/- 0.34 mm/h, RRD = 0.46, respectively. These paths show more balanced performance in rainfall-intensity stability, raindrop-spectrum realism, and perception consistency. The proposed method supports path selection, condition description, and credible interpretation of autonomous-driving perception tests under rainfall.

Problem

• simulated rainfall
• autonomous driving
• perception test
• rainfall credibility
• SOTIF

Innovation

• path-based evaluation
• equivalent rainfall intensity
• Realism of Raindrop Distribution (RRD)
• perception consistency
• simulated rainfall credibility