March 3, 2026Open Access

Learning From Real-World Trajectories: A Hybrid Genetic Algorithm and Reinforcement Learning Approach for Vulnerable Road User Model Calibration brochure

Key Points

Accurate calibration of the vulnerable road user model enhances real-world dynamics and interactions.
The hybrid approach combines genetic algorithms and reinforcement learning to improve calibration performance significantly.
Simulation techniques were analyzed to better reflect complex behaviors and interactions among road users in urban environments.
This approach may enable more realistic traffic modeling, though it requires further testing to confirm its efficacy.

Abstract

Microscopic simulation models require accurate calibration to reflect real-world dynamics. Vulnerable road user (VRU) model calibration is challenging due to: Complex interactions and behaviors, Mode-specific interactions, Diversity in infrastructure and the existence of shared spaces. Existing calibration methods, relying solely on numeric loss functions, often fail to capture realistic, human-perceived behavior.

Learning From Real-World Trajectories: A Hybrid Genetic Algorithm and Reinforcement Learning Approach for Vulnerable Road User Model Calibration brochure

Key Points

Abstract

Cite This Study