Simulating Automated Cars in a Virtual Urban Environment

Real world simulation by creating virtual environments is indispensable when you need to design complex entities which have to evolve in the real urban world. For example, testing control algorithms of autonomous vehicles needs an environment composed of a variety of active, moving entities and diierent kinds of vehicles. Virtual simulation environments have to provide this capability during the design step. To perform a realistic simulation of a real environment composed of a large set of vehicles (where some of them are autonomous, the others are controled by the user or by some speciic control law), we need to implement diierent models : geometric modeling of the environment, mechanical simulation of the vehicles, motion control models, driver models for autonomous vehicle, captor models to manage the interactions between objects, and visualisation algorithms. We present a simulation platform which complies with our needs.