Raw GIS to 3D road modeling for real-time traffic simulation

In this work, we propose a new approach to road modeling and 3D traffic simulation. Based on the raw geographic information system (GIS) data laid out as sparse polylines with attributes, we compute a more adequate functional description for real-time simulation of on-road vehicle animation. The proposed approach begins with a filtering/subdivision module where the raw polylines are transformed into a graph of functional road segments as arcs and the nodes as intersections. Then, the vehicle speed profile is computed based on its dynamics, its neighborhood and the curvature profile of the road. Afterward, a multi-agent system is proposed in order to handle a large number of simulated vehicle/driver couples. Finally, we deploy a 3D rendering engine to display the computed 3D simulation on screen. The resulting model satisfies most of the real road features for traffic simulation including road interchanges, roundabouts, intersections, lanes, etc. More importantly, the simulated driving qualitatively mimics the real behavior of the drivers/vehicles on the road as can be seen in the accompanying video (RTSP video). We also validate our findings with a technical assessment based on macroscopic and microscopic traffic simulation metrics in several road traffic scenarios.

[1]  Chen-Han Lee,et al.  An Improved B-Spline Fitting Method with Arc-Length Parameterization, G2-continuous Blending, and Quality Refinement , 2019, International Journal of Precision Engineering and Manufacturing.

[2]  A. Rababah,et al.  Approximating offset Curves using Bezier curves with high accuracy , 2020, International Journal of Electrical and Computer Engineering (IJECE).

[3]  Karthik K. Srinivasan,et al.  Study of traffic flow characteristics using different vehicle-following models under mixed traffic conditions , 2018 .

[4]  Abdenour Amamra,et al.  A GIS Data Realistic Road Generation Approach for Traffic Simulation , 2019, 2019 Federated Conference on Computer Science and Information Systems (FedCSIS).

[5]  Hideyuka Kita,et al.  EFFECTS OF MERGING LANE LENGTH ON THE MERGING BEHAVIOR AT EXPRESSWAY ON-RAMPS. , 1993 .

[6]  Ming C. Lin,et al.  Continuum Traffic Simulation , 2010, Comput. Graph. Forum.

[7]  Daniel Thalmann,et al.  An immersive multi-agent system for interactive applications , 2012, The Visual Computer.

[8]  Wolfram Burgard,et al.  Modeling and Planning Manipulation in Dynamic Environments , 2019, 2019 International Conference on Robotics and Automation (ICRA).

[9]  Daniël D. Heikoop,et al.  Driver Psychology during Automated Platooning , 2017 .

[10]  Meiyan Chen,et al.  Modeling and Simulation Analysis of Road Network Based on VISSIM , 2019, 2019 International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS).

[11]  Abdeslam El Fergougui,et al.  A Comparative Study of Urban Road Traffic Simulators , 2016 .

[12]  Swetava Ganguli,et al.  GeoGAN: A Conditional GAN with Reconstruction and Style Loss to Generate Standard Layer of Maps from Satellite Images , 2019, ArXiv.

[13]  Zhengfeng Ma,et al.  Model of the wheel motion state under path constraints based on the Darboux frame in 3D space , 2018, Engineering Computations.

[14]  Richard Romano,et al.  Optimal Path Planning with Clothoid Curves for Passenger Comfort , 2019, VEHITS.

[15]  Dirk Müller,et al.  Automated urban energy system modeling and thermal building simulation based on OpenStreetMap data sets , 2019, Building and Environment.

[16]  Fahui Wang,et al.  GIS-Based Simulation and Analysis of Intra-Urban Commuting , 2018 .

[17]  Marc Daniel,et al.  Realistic road path reconstruction from GIS data , 2014, Comput. Graph. Forum.

[18]  Ming C. Lin,et al.  City-scale traffic animation using statistical learning and metamodel-based optimization , 2017, ACM Trans. Graph..

[19]  Adam Millard-Ball,et al.  The world’s user-generated road map is more than 80% complete , 2017, PloS one.