Integrated Connected and Automated Vehicle Simulator for “InfoRich” Controls Project 2019-01-0676
Connected and automated vehicle (CAV) technologies will substantially decrease traffic accidents and traffic jams, and bring new opportunities for improving vehicle’s fuel economy. However, real world traffic is very complicated. It is costly to test an autonomous vehicle in real traffic environment, and real test scenario will not be able to cover various corner cases. It is also very challenging to create a controlled traffic environment that tests can be conducted repeatedly and compared fairly. Simulations are safer, more efficient, and cheaper than live testing on real vehicles. They also allow testing more scenarios than those that would be possible with real world testing. Therefore, it is important to develop a full-scale simulation platform, which can be used to simulate the traffic, vehicle, powertrain, even some critical components with certain level of details. In this work, a simulator is developed by integrating a 3D traffic modeling tool and a high fidelity vehicle model. The platform has a modular architecture, which supports hardware-in-the-loop (HiL) and vehicle-in-the-loop (ViL) testing. Customized sensor models, perception module and autonomous driving module can be implemented for ego vehicle. Smart and scalable traffic can be easily generated on the road network. This simulator can be used for testing how a vehicle would interact with other vehicles and infrastructure. 3D traffic scenarios can be created either from real road network or from statistically representative trip data. This platform contains tunable driver models which mimics human driver’s behavior. It can be used for developing control algorithm to improve CAV’s fuel economy in real world driving scenarios.