# Exploring Semantic Clustering and Similarity Search for Heterogeneous Traffic Scenario Graph

arXiv: https://arxiv.org/abs/2507.05086 · HTML: https://mlsanity.com/paper/2507.05086/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 22 references that are in the map.
- First posted on arXiv 7 Jul 2025.

## Builds on (most cited first, 18 of 18)

- [Graph Contrastive Learning with Augmentations](https://mlsanity.com/paper/2010.13902/) (2020)
- [Deep Graph Contrastive Representation Learning](https://mlsanity.com/paper/2006.04131/) (2020)
- [graph2vec: Learning Distributed Representations of Graphs](https://mlsanity.com/paper/1707.05005/) (2017)
- [Graph Matching Networks for Learning the Similarity of Graph Structured Objects](https://mlsanity.com/paper/1904.12787/) (2019)
- [Large-Scale Representation Learning on Graphs via Bootstrapping](https://mlsanity.com/paper/2102.06514/) (2021)
- [DriveGAN: Towards a Controllable High-Quality Neural Simulation](https://mlsanity.com/paper/2104.15060/) (2021)
- [HDGT: Heterogeneous Driving Graph Transformer for Multi-Agent Trajectory Prediction via Scene Encoding](https://mlsanity.com/paper/2205.09753/) (2022)
- [ScenarioNet: Open-Source Platform for Large-Scale Traffic Scenario Simulation and Modeling](https://mlsanity.com/paper/2306.12241/) (2023)
- [Generalizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated, and Tree](https://mlsanity.com/paper/1509.08985/) (2015)
- [Permutation-Invariant Variational Autoencoder for Graph-Level Representation Learning](https://mlsanity.com/paper/2104.09856/) (2021)
- [roadscene2vec: A Tool for Extracting and Embedding Road Scene-Graphs](https://mlsanity.com/paper/2109.01183/) (2021)
- [An Unsupervised Random Forest Clustering Technique for Automatic Traffic Scenario Categorization](https://mlsanity.com/paper/2004.02121/) (2020)
- [Towards Traffic Scene Description: The Semantic Scene Graph](https://mlsanity.com/paper/2111.10196/) (2021)
- [Traffic Scene Similarity: a Graph-based Contrastive Learning Approach](https://mlsanity.com/paper/2309.09720/) (2023)
- [VistaScenario: Interaction Scenario Engineering for Vehicles with Intelligent Systems for Transport Automation](https://mlsanity.com/paper/2402.07720/) (2024)
- [Geometric Deep Learning for Autonomous Driving: Unlocking the Power of Graph Neural Networks With CommonRoad-Geometric](https://mlsanity.com/paper/2302.01259/) (2023)
- [Heterogeneous Graph-based Trajectory Prediction using Local Map Context and Social Interactions](https://mlsanity.com/paper/2311.18553/) (2023)
- [Toward Unsupervised Test Scenario Extraction for Automated Driving Systems from Urban Naturalistic Road Traffic Data](https://mlsanity.com/paper/2202.06608/) (2022)

## Abstract (arXiv)

Scenario-based testing is an indispensable instrument for the comprehensive validation and verification of automated vehicles (AVs). However, finding a manageable and finite, yet representative subset of scenarios in a scalable, possibly unsupervised manner is notoriously challenging. Our work is meant to constitute a cornerstone to facilitate sample-efficient testing, while still capturing the diversity of relevant operational design domains (ODDs) and accounting for the "long tail" phenomenon in particular. To this end, we first propose an expressive and flexible heterogeneous, spatio-temporal graph model for representing traffic scenarios. Leveraging recent advances of graph neural networks (GNNs), we then propose a self-supervised method to learn a universal embedding space for scenario graphs that enables clustering and similarity search. In particular, we implement contrastive learning alongside a bootstrapping-based approach and evaluate their suitability for partitioning the scenario space. Experiments on the nuPlan dataset confirm the model's ability to capture semantics and thus group related scenarios in a meaningful way despite the absence of discrete class labels. Different scenario types materialize as distinct clusters. Our results demonstrate how variable-length traffic scenarios can be condensed into single vector representations that enable nearest-neighbor retrieval of representative candidates for distinct scenario categories. Notably, this is achieved without manual labeling or bias towards an explicit objective such as criticality. Ultimately, our approach can serve as a basis for scalable selection of scenarios to further enhance the efficiency and robustness of testing AVs in simulation.