A GPU-based Method for Massive Simulation of Distributed Behavioral Models with CUDA

This work reports the results of a GPU-based approach for the massive simulation of a dis- tributed behavioral model. In this model an agent has a local perception of the world and then it moves by coordinating with the motion of its neighbors. This carries a very high com- putational cost in the so-called nearest neighbors search. By leveraging the parallel processing power of the GPU and its new programming model called CUDA, we implemented a spatial hashing where a partitioning of the space is used to accelerate the neighbors search. Through extensive experiments, we demonstrate the eectiveness of our GPU implementation when simulating the motion of high-density agent groups