The Research About The Basic Algorithm Of Crowd Movement Based On Fluid Simulation

With the development of computer technology and crowd simulation,massive crowd simulation has become a new hot topical and much higher demands of realistic,real-time and efficiency are needed in crowd simulation.Therefore,studies on massive crowd simulation are of both scientific and application values In order to simulate crowd motion efficiently and veritably,an algorithm based on fluid dynamics was developed to stimulate the movement of massive crowd.This method can reproduce large crowd virtual agents and provide users with simple and flexible control tools.This simulation algorithm focus on controlling the movement of crowd and the controlling strategy allows real-time user control and simulation rate.A method for simulation the behaviors of large crowds based on radial basis functions(RBF)is studied.Inspired by RBF method,we employ a global control theme to analyze the behavior of large crowd in real-time.A velocity field covering the whole simulation area is firstly created,by which all individuals will be driven.Differently from RBF method,the smooth vector field is obtained by assuming the crowds flow as fluid flow and solving the Navier-Stokes(NS)equations which describes 2D incompressible fluid.During the simulation,the velocity field can be changed by adjusting the parameters of the Navier-Stokes equation,thus changing the movement of the crowds.Additionally,Based on smoothed particle hydrodynamics,a design approach of collisionavoidance and detection is proposed.Smoothed particle hydrodynamics will be used tosimulate a possible collision to detection and avoid in the crowd without using the bounding box.It will improve the efficiency of system simulation.In the experiment part,the algorithms of RBF simulation,Treuille potential field simulation and Narain simulation were compared.The results showed that the proposed algorithm is successful in massive crowd simulation and it shows great potential in further applications in complex environments.This method gives excellent results of pedestrians movement.