Deep Modeling of Non-Gaussian Aleatoric Uncertainty

被引：0

作者：

Acharya, Aastha ^{[1
,2
]}

Lee, Caleb ^{[1
]}

D'Alonzo, Marissa ^{[1
]}

Shamwell, Jared ^{[1
]}

Ahmed, Nisar R. ^{[2
]}

Russell, Rebecca ^{[1
]}

机构：

[1] Charles Stark Draper Lab Inc, Cambridge, MA 02139 USA

[2] Univ Colorado Boulder, Ann & H J Smead Dept Aerosp Engn Sci, Boulder, CO 80303 USA

来源：

IEEE ROBOTICS AND AUTOMATION LETTERS | 2025年 / 10卷 / 01期

关键词：

Uncertainty; Estimation; Predictive models; Robot sensing systems; Deep learning; Probability density function; State estimation; Navigation; Data models; Gaussian distribution; Deep learning methods; deep learning for visual perception; vision-based navigation;

D O I：

10.1109/LRA.2024.3511376

中图分类号：

TP24 [机器人技术];

学科分类号：

080202 ; 1405 ;

摘要：

Deep learning offers promising new ways to accurately model aleatoric uncertainty in robotic state estimation systems, particularly when the uncertainty distributions do not conform to traditional assumptions of being fixed and Gaussian. In this study, we formulate and evaluate three fundamental deep learning approaches for conditional probability density modeling to quantify non-Gaussian aleatoric uncertainty: parametric, discretized, and generative modeling. We systematically compare the respective strengths and weaknesses of these three methods on simulated non-Gaussian densities as well as on real-world terrain-relative navigation data. Our results show that these deep learning methods can accurately capture complex uncertainty patterns, highlighting their potential for improving the reliability and robustness of estimation systems.

引用

页码：660 / 667

页数：8

共 35 条

[1] Learning to Forecast Aleatoric and Epistemic Uncertainties over Long Horizon Trajectories [J].