Handling of Contacts in Crowd Motion Simulations

This paper is concerned with the modelling of crowd motion in highly packed configurations. We are especially interested in describing evacuation situations: people want to escape from a room, a building, a railway station or a plane, that may contain obstacles (walls, seats, tables, ...). Many strategies have been followed to model crowd motions: macroscopic models [14, 9, 15], Cellular Automata [2, 16, 3], queuing models [17, 23] and microscopic models [8, 11]. The possibility that people may actually get into contact is usually handled indirectly (repulsive forces or adapted cut-off for microscopic models, exclusion principle for Cellular Automata or queuing models). We propose here to integrate a strong non-overlapping constraint in a ODE framework, in the spirit of granular flow models. Our approach rests on two principles. On the one hand, we define a spontaneous velocity, which corresponds to the velocity each individual would like to have in the absence of other people. On the other hand, individuals (which are identified to rigid discs) must obey a non-overlapping constraint. Those two principles lead us to define the actual velocity field as the projection of the spontaneous velocity over the set of admissible velocities (regarding the nonoverlapping constraints). To perform this projection, we put the problem in a saddle-point form, which leads us to introduce a collection of Lagrange multipliers. Those Lagrange multipliers can be interpreted as interaction pressure between people which are in contact.

[1]  J. MacGregor Smith,et al.  Modeling circulation systems in buildings using state dependent queueing models , 1989, Queueing Syst. Theory Appl..

[2]  Helbing,et al.  Social force model for pedestrian dynamics. , 1995, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[3]  D. Helbing,et al.  Lattice gas simulation of experimentally studied evacuation dynamics. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[4]  Serge P. Hoogendoorn,et al.  DYNAMIC USER-OPTIMAL ASSIGNMENT IN CONTINUOUS TIME AND SPACE , 2004 .

[5]  Victor J. Blue,et al.  Cellular automata microsimulation for modeling bi-directional pedestrian walkways , 2001 .

[6]  P. G. Ciarlet,et al.  Introduction a l'analyse numerique matricielle et a l'optimisation , 1984 .

[7]  P G Gipps,et al.  A micro simulation model for pedestrian flows , 1985 .

[8]  Dirk Helbing A Fluid-Dynamic Model for the Movement of Pedestrians , 1992, Complex Syst..

[9]  L. Thibault Sweeping process with regular and nonregular sets , 2003 .

[10]  Gunnar G. Løvås,et al.  Modeling and Simulation of Pedestrian Traffic Flow , 1994 .

[11]  H. Brezis Opérateurs maximaux monotones et semi-groupes de contractions dans les espaces de Hilbert , 1973 .

[12]  Lionel Thibault,et al.  BV solutions of nonconvex sweeping process differential inclusion with perturbation , 2006 .

[13]  Michael Schreckenberg,et al.  Pedestrian and evacuation dynamics , 2002 .

[14]  Giovanni Colombo,et al.  Sweeping by a continuous prox-regular set $ , 2003 .

[15]  Dirk Helbing,et al.  Simulating dynamical features of escape panic , 2000, Nature.

[16]  Bertrand Maury,et al.  A time-stepping scheme for inelastic collisions , 2006, Numerische Mathematik.

[17]  Andreas Schadschneider,et al.  Simulation of evacuation processes using a bionics-inspired cellular automaton model for pedestrian dynamics , 2002 .

[18]  A. Schadschneider,et al.  Simulation of pedestrian dynamics using a two dimensional cellular automaton , 2001 .

[19]  L. F. Henderson On the fluid mechanics of human crowd motion , 1974 .

[20]  Dirk Helbing,et al.  Optimal self-organization , 1999 .

[21]  J. Moreau Evolution problem associated with a moving convex set in a Hilbert space , 1977 .

[22]  Lionel Thibault,et al.  Relaxation of an optimal control problem involving a perturbed sweeping process , 2005, Math. Program..