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

Details

• Corporate Creators:
  United States. Department of Transportation. Federal Highway Administration
• Corporate Contributors:
  United States. Department of Transportation. Federal Highway Administration. Turner-Fairbank Highway Research Center
• Subject/TRT Terms:
  • Vehicle trajectories
  • Simulation
  • Vulnerable road users
  • Genetic algorithms
  • Learning
• Publication/ Report Number:
  FHWA-HRT-26-033
• DOI:
  https://doi.org/10.21949/1qgs-s409
• Resource Type:
  Booklet/Pamphlet
• Right Statement:
  Public Domain
• Geographical Coverage:
  United States
• Corporate Publisher:
  United States. Department of Transportation. Federal Highway Administration
• 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.
• Format:
  PDF
• Funding:
  693JJ324C000004
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/88973/dot_88973_DS1.pdf
• File Type:
  [PDF - 864.34 KB]
• Collection(s):
  Artificial Intelligence Deployment Center of Excellence (AI COE) ; Federal Highway Administration
• Main Document Checksum:
  urn:sha-512:343240213b80d58b1b55c47ac10ca6b50667a25c447d97e8bbb14c7f0239414d02e762f312aae21829aceb13dfb5a54b8b3b3a4d1a646ca0519f6d2a3cc646ff