Probabilistic Rainfall Estimation from Automotive Lidar

TL;DR

This paper introduces a probabilistic hierarchical Bayesian model that accurately estimates rainfall rates from automotive lidar data, enhancing autonomous vehicle perception in adverse weather conditions.

Contribution

It presents a novel probabilistic decision tree model for rainfall estimation from lidar, validated with large-scale experimental data, demonstrating high accuracy and sensor independence.

Findings

Prediction accuracy comparable to disdrometer resolution

RMSE of 2.42 mm/h after filtering uncertain predictions

Model performance varies with tree depth and sampling parameters

Abstract

Robust sensing and perception in adverse weather conditions remain one of the biggest challenges for realizing reliable autonomous vehicle mobility services. Prior work has established that rainfall rate is a useful measure for the adversity of atmospheric weather conditions. This work presents a probabilistic hierarchical Bayesian model that infers rainfall rate from automotive lidar point cloud sequences with high accuracy and reliability. The model is a hierarchical mixture of experts model, or a probabilistic decision tree, with gating and expert nodes consisting of variational logistic and linear regression models. Experimental data used to train and evaluate the model is collected in a large-scale rainfall experiment facility from both stationary and moving vehicle platforms. The results show prediction accuracy comparable to the measurement resolution of a disdrometer, and the soundness and usefulness of the uncertainty estimation. The model achieves RMSE 2.42\,mm/h after filtering out uncertain predictions. The error is comparable to the mean rainfall rate change of 3.5\,mm/h between measurements. Model parameter studies show how predictive performance changes with tree depth, sampling duration, and crop box dimension. A second experiment demonstrates the predictability of higher rainfall above 300\,mm/h using a different lidar sensor, demonstrating sensor independence.