02:00 PM - 03:00 PM
Benjamin Hernandez, Barcelona Supercomputing Center, Barcelona, Spain
National Center for Computational Sciences Seminar
Research Office Building (5700), Room L-204
Email: T. P. StraatsmaPhone:
The use of computational models that describe the behavior of people is becoming a necessity as the population increases in urban areas. Prediction before, during and after daily crowd events may reduce associated logistic costs. In this sense, large scale crowd simulations demand computational power and memory resources commonly available in HPC platforms.
In this talk, I will present a system architecture to achieve real-time crowd simulation and visualization. The engine is composed by three modules: the crowd simulation module controls agent’s path planning, collision avoidance and response. The visualization module generates and renders detailed 3D characters, and the streaming module enables different configurations of the system:
1. Streaming Mode. The simulation runs (or previously stored) in a remote resource, and results are streamed to a visualization server.
2. In-situ Mode. The simulation and visualization run in a GPU cluster, each node execute both simulation and visualization of one tile of the world.
3. Web-view mode. This mode is designed to allow the researcher accessing visualization results in any device and away from the desktop.
In addition I will briefly describe other use cases of the proposed architecture in Molecular and Active Particles simulations.
About the Speaker
Dr. Hernandez is a postdoctoral researcher at Barcelona Supercomputing Center in Spain. Previously he was full time professor at Tecnologico de Monterrey Campus Ciudad de Mexico, Mexico from 2010 to 2012. As a professor he has advised postgraduate thesis on real-time crowd and traffic simulation, and human computer interaction and co-directed the NVIDIA CUDA Teaching Center Initiative from 2010 to 2013. His research experience also includes internships at Microsoft Research’s Graphics Group in Redmond, WA and Trinity College’s GV2 Lab in Dublin, Ireland. He currently holds the National Research System Fellowship from the Mexican Research Council (CONACyT), México.
Dr. Hernández focuses his research interests at the intersection of real time crowd simulation, human computer interaction and visualization of inhabited virtual environments using high performance computing. He has published research papers on these topics in international journals, conference venues such as Siggraph, Conference on Computer Animation and Social Agents (CASA), Motion In Games (MIG), Computer Graphics International (CGI), and book chapters such as Shader X and GPU Pro book series.