Cooperative Ramp Merging System: Agent-Based Modeling and Simulation Using Game Engine
- Kyungtae Han - Toyota Motor North America, USA ,
- BaekGyu Kim - Toyota Motor North America, USA ,
- Prashant Tiwari - Toyota Motor North America, USA ,
- Ziran Wang - University of California, Riverside, USA ,
- Guoyuan Wu - University of California, Riverside, USA ,
- Kanok Boriboonsomsin - University of California, Riverside, USA ,
- Matthew J. Barth - University of California, Riverside, USA
ISSN: 2574-0741, e-ISSN: 2574-075X
Published May 16, 2019 by SAE International in United States
Citation: Wang, Z., Wu, G., Boriboonsomsin, K., Barth, M. et al., "Cooperative Ramp Merging System: Agent-Based Modeling and Simulation Using Game Engine," SAE Intl. J CAV 2(2):115-128, 2019, https://doi.org/10.4271/12-02-02-0008.
Agent-based modeling and simulation (ABMS) has been a popular approach for modeling autonomous and interacting agents in a multi-agent system. Specifically, ABMS can be applied to connected and automated vehicles (CAVs) since CAVs can operate autonomously with the help of onboard sensors, and cooperate with each other through vehicle-to-everything (V2X) communications. In order to improve energy efficiency and mobility of traffic, we have developed an online feedforward/feedback longitudinal controller for CAVs to cooperatively merge at ramps. Agent-based CAV models were built in the Unity3D environment, where vehicles are given connectivity and autonomy through C#-based scripting application programming interface (API). Agent-based infrastructure model is also built as a Unity3D simulation network based on the city of Mountain View, California. A simulation of cooperative on-ramp merging is carried out with a distributed consensus-based protocol, and then compared with the human-in-the-loop simulation where the on-ramp merging vehicle is driven by four different human drivers on a driving simulator. The benefits of introducing the proposed protocol are evaluated in terms of travel time, energy consumption, and pollutant emissions. The results show that the proposed cooperative on-ramp merging protocol can reduce average travel time, energy consumption, and pollutant emissions by 7%, 8%, and 58%, respectively, when compared to the human-in-the-loop scenario.