Development of a Mixed Traffic Simulation Environment with Heavy Vehicles Incorporating a Merging Decision Model Based on Empirical Trajectories

Kenta Shintoku･Nobuhiro Araki･Tomoyo Saitou･Tushar Shrotriya (Kozo Keikaku Engineering)･Hiroyuki Furushou･Hisashi Imanaga･Hideo Nakamura (JARI)･Hiroshi Fujimoto (JAMA)

This study extracted and modeled the human merging decision-making process from empirical trajectories to evaluate natural merging behaviors in mixed traffic flows containing automated and advanced driver-assistance vehicles. Subsequently, by integrating both the mixing of heavy vehicles with distinct characteristics and a merging scene classification into the traffic flow simulation, the high-accuracy reproducibility of vehicle behaviors under actual conditions was confirmed.

Evaluation of merging support system using assumed automated driving vehicles on main lane of Highway
-(First Report) Effective penetration rate of automated driving vehicles in the system at early penetration stage-

Hisashi Imanaga (JARI)･Kenta Shintoku･Nobuhiro Araki･Tushar Shrotriya･Ryota Okazaki (Kozo Keikaku Engineering)･Hideo Nakamura (JARI)･Hiroshi Fujimoto (JAMA)

This paper proposes a merging support system assuming the widespread use of automated vehicles on highway main lanes. It involves increasing the distance between leading vehicles to facilitate merging. Traffic flow simulations confirmed that, assuming a low proportion of automated vehicles in the initial stages of adoption, this system can minimize negative impacts up to a 30% adoption rate of automated vehicles.

Evaluating Traffic Control in Pedestrian-Vehicle Mixed Spaces Using Agent-Based Simulation Considering Personal Characteristics

Shunto Araki･Hiroyuki Okuda (Nagoya University)･Kazunori Ban (Toyota Technical Development)･Tatsuya Suzuki (Nagoya University)

In pedestrian-vehicle mixed spaces, it is important to consider the impact of traffic participants' personal decision-making characteristics on traffic flow. In this study, we probabilistically modeled individual decision-making characteristics based on experimental data collected in a virtual environment, and constructed an agent-based simulation incorporating these models. Using the constructed simulation, we execute various traffic control scenarios to evaluate traffic control approaches from the perspectives of efficiency and safety.

Development of an Optimal Charging Infrastructure Deployment Model Considering the Adoption of Electric Trucks for Long-Distance Transport

Soichiro Ishii･Hirohisa Aki (University of Tsukuba)

In this study, we developed an optimal deployment model for charging infrastructure that combines dynamic wireless power transfer systems and quick charging facilities at parking areas, targeting electric trucks used for long-distance transport on expressways. Furthermore, we performed traffic simulations based on the vehicle penetration status to analyze facility deployment, charging facility use, and the impact on power load.