Integrating Human Behavior Toward the Development of Safer Cooperative Automated Transportation: Implementation of SHRP2 Naturalistic Driving Study

Ahmed, Mohamed; Das, Anik; Khan, Md Nasim; Sigdel, Mandip; Maddineni, Vamsi · 2023 · ROSA P / Wyoming. Department of Transportation

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study addresses the challenges of integrating Cooperative Automated Transportation (CAT)—encompassing Connected Vehicles (CV), Autonomous Vehicles (AV), and Connected and Automated Vehicles (CAV)—into existing infrastructure with mixed traffic. The primary motivation is to understand human driving behavior, specifically lane-changing maneuvers, to ensure safety and operational efficiency in environments where human-driven vehicles interact with automated systems. The research utilizes data from the Second Strategic Highway Research Program (SHRP2) Naturalistic Driving Study (NDS) to develop and validate microsimulation models that assess CAT performance under varying Market Penetration Rates (MPRs) and Levels of Autonomy. The methodology combines deep learning techniques with traffic microsimulation. First, the authors developed a ResNet-18 Convolutional Neural Network (CNN) to detect lane-change maneuvers from SHRP2 NDS data. They evaluated feature importance using Boruta and XGBoost across six data categories, including vehicle kinematics, machine vision, roadway geometry, and driver demographics. The study found that models incorporating vehicle kinematics and machine vision features yielded the highest detection accuracy. These insights were used to update and optimize lane-change parameters in the PTV VISSIM microsimulation platform, specifically adjusting for adverse weather conditions such as heavy rain. The calibrated models were then applied to simulate freeway segments, including weaving and basic segments, as well as work zones in Wyoming. The results demonstrate that integrating human behavior data significantly improves the accuracy of safety and operational assessments. In freeway simulations, the study analyzed traffic conflicts using surrogate measures of safety (SMoS) such as Time to Collision (TTC), Post Encroachment Time (PET), and Deceleration Rate to Avoid Collision (DRAC). Findings indicated that increasing MPRs generally reduced conflict frequencies and improved operational metrics like average speed and total travel time, though the impact varied by automation level and infrastructure type. In work zone scenarios, the introduction of Cooperative Adaptive Cruise Control (CACC) truck platoons was shown to reduce queue lengths and conflict frequencies, particularly when specific lane configurations were optimized for platoon control during lane closures. The study also highlighted that adverse weather conditions necessitated distinct parameter adjustments to maintain safety margins in mixed traffic. The significance of this work lies in its provision of a validated framework for modeling mixed traffic environments using real-world naturalistic driving data. By linking deep learning-based behavior detection with microsimulation, the study offers actionable insights for transportation agencies on how CAT adoption affects safety and operations. The findings support the development of safer cooperative automated transportation systems by identifying optimal lane configurations, platoon strategies, and parameter adjustments required for different infrastructure contexts and weather conditions. This approach aids in predicting the benefits of CAT deployment and informs infrastructure planning to accommodate varying levels of vehicle automation and connectivity.

Key finding

Integrating SHRP2 naturalistic driving data into microsimulation models via deep learning allows for more accurate assessment of safety and operational impacts of cooperative automated transportation under varying market penetration rates and adverse weather conditions.

Methodology

mixed_methods

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• traffic density
• work zones
• situational awareness
• naturalistic crash near crash
• lane changing
• speed choice

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Methodological Resource: dataset resource, tool software
• Theoretical Contribution: computational model