High-Fidelity Roadway Modeling and Simulation

Roads are an essential feature in our daily lives. With the advances in computing technologies, 2D and 3D road models are employed in many applications, such as computer games and virtual environments. Traditional road models were generated by professional artists manually using modeling software tools such as Maya and 3ds Max. This approach requires both highly specialized and sophisticated skills and massive manual labor. Automatic road generation based on procedural modeling can create road models using specially designed computer algorithms or procedures, reducing the tedious manual editing needed for road modeling dramatically. But most existing procedural modeling methods for road generation put emphasis on the visual effects of the generated roads, not the geometrical and architectural fidelity. This limitation seriously restricts the applicability of the generated road models. To address this problem, this paper proposes a high-fidelity roadway generation method that takes into account road design principles practiced by civil engineering professionals, and as a result, the generated roads can support not only general applications such as games and simulations in which roads are used as 3D assets, but also demanding civil engineering applications, which requires accurate geometrical models of roads. The inputs to the proposed method include road specifications, civil engineering road design rules, terrain information, and surrounding environment. Then the proposed method generates in real time 3D roads that have both high visual and geometrical fidelities. This paper discusses in details the procedures that convert 2D roads specified in shape files into 3D roads and civil engineering road design principles. The proposed method can be used in many applications that have stringent requirements on high precision 3D models, such as driving simulations and road design prototyping. Preliminary results demonstrate the effectiveness of the proposed method.