Agent-based simulation of collective cooperation: from experiment to model

Benedikt Kleinmeier,John Drury,Gerta Köster

doi:10.1098/rsif.2020.0396

Abstract

Simulation models of pedestrian dynamics have become an invaluable tool for evacuation planning. Typically, crowds are assumed to stream unidirectionally towards a safe area. Simulated agents avoid collisions through mechanisms that belong to each individual, such as being repelled from each other by imaginary forces. But classic locomotion models fail when collective cooperation is called for, notably when an agent, say a first-aid attendant, needs to forge a path through a densely packed group. We present a controlled experiment to observe what happens when humans pass through a dense static crowd. We formulate and test hypotheses on salient phenomena. We discuss our observations in a psychological framework. We derive a model that incorporates: agents’ perception and cognitive processing of a situation that needs cooperation; selection from a portfolio of behaviours, such as being cooperative; and a suitable action, such as swapping places. Agents’ ability to successfully get through a dense crowd emerges as an effect of the psychological model.

Highlights

Simulation models of pedestrian dynamics are widely used today especially for evacuation planning [1,2,3,4]
— Pedestrians walking through a crowd are slowed down. — The pedestrians in a waiting crowd return to their initial positions after giving way to the ‘intruder’. — Real humans can pass a crowd at high densities
We identified a major shortcoming in current pedestrian simulation models: the lack of collective cooperation which means that agents fail at seemingly simple tasks, such as forging a path through a dense crowd

Summary

Introduction

Simulation models of pedestrian dynamics are widely used today especially for evacuation planning [1,2,3,4] Such models usually consist of unidirectional flows of agents (simulated pedestrians) and are used to estimate the evacuation time in emergency situations or to test safety concepts [5]. Simulations of such models are a useful tool in the planning phase to detect critical high densities for example to avoid casualties such as those reported at the Hajj on several occasions [6, p. When reenacting such a real-world situation in current simulation tools, agents often get stuck and end up in a deadlock situation because there is no real interaction between agents, compare figure 1

Related work

Goals of our work and article structure

Experiment set-up

Experiment results

Experiment result: speed distributions

Experiment result: distribution of the waiting crowd

The need for a psychology model complementing pure locomotion

Model of a psychology layer for collective cooperation

Extend the existing simulation loop

Re-enacting the experiment with the collective cooperation model

Conclusion