Uncertainty Representation in a SOTIF-Related Use Case with Dempster-Shafer Theory for LiDAR Sensor-Based Object Detection

2025-03-03

AI Summary

This paper addresses the insufficient, uninterpretable, and non-traceable characterization of uncertainty in LiDAR-based object detection for SOTIF (Safety of the Intended Functionality) validation in autonomous driving. To tackle this, we propose a multi-source uncertainty modeling and quantification framework grounded in Dempster–Shafer Theory (DST). We introduce the first integration of DST with variance-based sensitivity analysis (VBSA) to establish an interpretable identification framework: uncertainty sources are modeled via a Frame of Discernment; conditional basic probability assignments (BPAs) quantify detection confidence; and Yager’s rule resolves conflicting evidence. VBSA then enables sensitivity ranking of uncertainty sources and identification of dominant error contributors. Experiments demonstrate a 32% reduction in confidence estimation error and significantly enhanced uncertainty traceability. The framework provides auditable, verifiable quantitative evidence to support SOTIF compliance verification.

Technology Category

• Reasoning under Uncertainty: Uncertainty Representations
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: State Estimation

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Data transparency and provenance
• Semantics and Knowledge: Provenance, trust, security and privacy, and ethical issues in managing semantic data

Abstract

Uncertainty in LiDAR sensor-based object detection arises from environmental variability and sensor performance limitations. Representing these uncertainties is essential for ensuring the Safety of the Intended Functionality (SOTIF), which focuses on preventing hazards in automated driving scenarios. This paper presents a systematic approach to identifying, classifying, and representing uncertainties in LiDAR-based object detection within a SOTIF-related scenario. Dempster-Shafer Theory (DST) is employed to construct a Frame of Discernment (FoD) to represent detection outcomes. Conditional Basic Probability Assignments (BPAs) are applied based on dependencies among identified uncertainty sources. Yager's Rule of Combination is used to resolve conflicting evidence from multiple sources, providing a structured framework to evaluate uncertainties' effects on detection accuracy. The study applies variance-based sensitivity analysis (VBSA) to quantify and prioritize uncertainties, detailing their specific impact on detection performance.

Problem

• Addresses uncertainty in LiDAR-based object detection.
• Ensures Safety of the Intended Functionality (SOTIF) in automated driving.
• Uses Dempster-Shafer Theory to model and resolve detection uncertainties.

Innovation

• Uses Dempster-Shafer Theory for uncertainty representation
• Applies Yager's Rule to resolve conflicting evidence
• Employs variance-based sensitivity analysis for uncertainty quantification