Finkeldei, Florian (Technische Universität München), Wolf, Michael (TUM), Weghorn, Jan-Niklas (Technical University of Munich), Pretschner, Alexander (Technical University of Munich), Althoff, Matthias (Technische Universität München)

Enhanced Traffic Rule Monitoring Using Model-Predictive and Duration-Aware Robustness

Scheduled for presentation during the Regular Session "S27a-Safety and Risk Assessment for Autonomous Driving Systems" (TH-LM-T27), Thursday, November 20, 2025, 11:30−11:50, Broadbeach 3

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on October 18, 2025

Abstract

Monitoring traffic rule compliance is essential for automated driving. To steer motion planning towards rule-compliant solutions, the degree of traffic rule violation is indispensable – this is often referred to as the robustness of traffic rule compliance. Existing approaches for robustness monitoring of temporal logics suffer from inaccuracies in determining the robustness of individual predicates and aggregating their robustness. We address the inaccurate robustness of predicates by using model-predictive robustness and improve aggregating robustness by introducing two novel robustness definitions for temporal operators. These robustness definitions consider the duration of specification violations in addition to their severity. The novel robustness definitions fulfill relevant properties, such as soundness, boundedness, and monotonicity. In numerical experiments, we demonstrate that model-predictive robustness has an improved sensitivity and show the benefits of duration-aware robustness.