A simulation platform calibration method for automated vehicle evaluation: accurate on both vehicle level and traffic flow level

2025-12-17

AI Summary

Existing traffic simulation platforms struggle to simultaneously achieve high-fidelity calibration of autonomous vehicles (AVs) and background traffic at both the microscopic (vehicle-to-vehicle interaction) and macroscopic (traffic flow) levels. To address this, we propose a novel two-tier collaborative calibration paradigm specifically designed for AV evaluation. Our method introduces the first end-to-end framework supporting explicit vehicle interaction modeling, fully automated pipeline execution, and multi-scale optimization integration. Leveraging interaction-driven modeling and hierarchical accuracy constraints, it concurrently ensures fidelity in both individual vehicle behavior and aggregate traffic flow metrics. Experimental results demonstrate an 83.53% improvement in interaction reproduction accuracy, a 76.75% increase in calibration efficiency, and a 51.9% gain in overall accuracy across both micro- and macro-scale performance indicators—all without manual intervention. This work establishes a scalable, fully automated calibration methodology for high-fidelity traffic simulation.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

Simulation testing is a fundamental approach for evaluating automated vehicles (AVs). To ensure its reliability, it is crucial to accurately replicate interactions between AVs and background traffic, which necessitates effective calibration. However, existing calibration methods often fall short in achieving this goal. To address this gap, this study introduces a simulation platform calibration method that ensures high accuracy at both the vehicle and traffic flow levels. The method offers several key features:(1) with the capability of calibration for vehicle-to-vehicle interaction; (2) with accuracy assurance; (3) with enhanced efficiency; (4) with pipeline calibration capability. The proposed method is benchmarked against a baseline with no calibration and a state-of-the-art calibration method. Results show that it enhances the accuracy of interaction replication by 83.53% and boosts calibration efficiency by 76.75%. Furthermore, it maintains accuracy across both vehicle-level and traffic flow-level metrics, with an improvement of 51.9%. Notably, the entire calibration process is fully automated, requiring no human intervention.

Problem

• Calibrates simulation platforms for automated vehicle evaluation
• Ensures accuracy at both vehicle and traffic flow levels
• Automates calibration to improve interaction replication and efficiency

Innovation

• Calibrates vehicle-to-vehicle interaction automatically
• Ensures accuracy at both vehicle and traffic flow levels
• Improves calibration efficiency by over 75%