Research map / Large language models / Agents, code and math

Traffic Scenario Logic: A Spatial-Temporal Logic for Modeling and Reasoning of Urban Traffic Scenarios

AAAI 2025 · Large language models › Agents, code and math · approach: Reasoning benchmarks and platforms

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Traffic Scenario Logic, a spatial-temporal logic for modeling and reasoning about urban pedestrian-free traffic scenarios derived from OpenDRIVE road networks, implemented with Telingo (ASP) and demonstrated on several road layouts.

Problem

Existing formal representations of traffic scenarios are limited to highways or simplified intersections and rely on discretized road/grid representations, which hinders generating test cases for urban autonomous driving.

Approach

A spatial-temporal logic that represents vehicle longitudinal position relative to other vehicles and to road points (intersections, connections, overlaps) instead of discretized positions, with a road network abstraction derivable from OpenDRIVE, and rules encoded as temporal ASP constraints for scenario generation.

Key results

• TSL is demonstrated via reasoning examples (highway overtake, intersection, connecting roads, vehicle passing two intersections, opposite-direction overtake) that generate multiple distinct scenarios on different urban road layouts.
• The paper claims a formal representation of OpenDRIVE urban road networks without discretization approximations, with a theorem asserting existence of such a representation for structured 2D OpenDRIVE networks.
• Translation of TSL scenarios into OpenSCENARIO DSL is described as a straightforward, automatable procedure.

Limitations

• Restricted to pedestrian-free traffic scenarios; pedestrians left for future work.
• Evaluation is limited to qualitative illustrative examples rather than quantitative benchmarks.
• Safety verification of control sequences and downstream simulator use are not evaluated in the paper.

Experimental setup

Builds on: Multi-Lane Spatial Logic Urban Multi-Lane Spatial Logic Telingo OpenDRIVE OpenSCENARIO DSL

Closest papers by meaning

• DriveCombo: Benchmarking Compositional Traffic Rule Reasoning in Autonomous Driving (CVPR 2026 · similarity 0.88)
• STSBench: A Spatio-temporal Scenario Benchmark for Multi-modal Large Language Models in Autonomous Driving (NeurIPS 2025 · similarity 0.86)
• VLMLight: Safety-Critical Traffic Signal Control via Vision-Language Meta-Control and Dual-Branch Reasoning Architecture (NeurIPS 2025 · similarity 0.86)
• T2SG: Traffic Topology Scene Graph for Topology Reasoning in Autonomous Driving (CVPR 2025 · similarity 0.85)
• Hybrid-Driving: An Autonomous Driving Decision Framework Integrating Large Language Models, Knowledge Graphs and Driving Rules (AAAI 2025 · similarity 0.84)
• TopoMLP: A Simple yet Strong Pipeline for Driving Topology Reasoning (ICLR 2024 · similarity 0.84)