GPU Accelerated Nature Inspired Methods for Modelling Large Scale Bi-directional Pedestrian Movement

Pedestrian movement, although ubiquitous and well-studied, is still not that well understood due to the complicating nature of the embedded social dynamics. Interest among researchers in simulating pedestrian movement and interactions has grown significantly in part due to increased computational and visualization capabilities afforded by high power computing. Different approaches have been adopted to simulate pedestrian movement under various circumstances and interactions. In the present work, bi-directional crowd movement is simulated where an equal numbers of individuals try to reach the opposite sides of an environment. Two movement methods are considered. First a Least Effort Model (LEM) is investigated where agents try to take an optimal path with as minimal changes from their intended path as possible. Following this, a modified form of Ant Colony Optimization (ACO) is proposed, where individuals are guided by a goal of reaching the other side in a least effort mode as well as a pheromone trail left by predecessors. The basic idea is to increase agent interaction, thereby more closely reflecting a real world scenario. The methodology utilizes Graphics Processing Units (GPUs) for general purpose computing using the CUDA platform. Because of the inherent parallel properties associated with pedestrian movement such as proximate interactions of individuals on a 2D grid, GPUs are well suited. The main feature of the implementation undertaken here is that the parallelism is data driven. The data driven implementation leads to a speedup up to 18x compared to its sequential counterpart running on a single threaded CPU. The numbers of pedestrians considered in the model ranged from 2K to 100K representing numbers typical of mass gathering events. A detailed discussion addresses implementation challenges faced and averted. Detailed analysis is also provided on the throughput of pedestrians across the environment.