Mode of access: World Wide Web.

System requirements: Adobe Acrobat reader.

Part of: Synthesis digital library of engineering and computer science.

Series from website.

Includes bibliographical references (p. 163-[174]).

Introduction -- Terminology -- Overview -- Lessons learned from the psychology literature -- Main features in crowd simulation systems -- Crowd simulation methodology survey -- Microscopic and macroscopic approaches used to model pedestrian movements -- Microscopic models -- Social force models -- Cellular automata models -- Rule-based models -- Macroscopic models -- Regression models -- Route choice models -- Queuing models -- Gaskinetics -- Current pedestrian software systems -- Summary of crowd models -- Some limitations of the current commercial software for crowd evacuation -- Navigation -- Cell and portal graphs -- Flow tiles and potential field methods -- Probabilistic roadmaps -- Environment modeling -- Individual differences in crowds -- Personality and emotion models -- Physiology -- Sociology: subgroups -- Culture, roles, and status -- Summary -- Framework (HiDAC + MACES + CAROSA) -- Interaction between framework levels and psychological models -- Parameters affecting crowd behavior -- HiDAC local motion -- Introduction -- Agents' speeds and densities -- Walking speeds and densities when walking downstairs -- Perception -- Crossing portals -- The HiDAC model -- Avoidance forces -- Repulsion forces -- Solution to "shaking" problem in high densities -- Organized behavior: queuing -- Pushing behavior -- Falling and becoming obstacles -- Panic propagation -- MACES: wayfinding with communication and roles -- Introduction -- Navigation algorithm -- Exploring the building -- Communication affecting evacuation times -- Relevance of having trained leaders vs. untrained leaders -- Importance of leadership -- Simulating psychology affecting roles and navigation -- Interactive navigation and impatient agents avoiding bottlenecks -- CAROSA: functional crowds -- Applications with actions -- Parameterized action representation -- Key fields of the action representation -- Key fields of the object representation -- Four types of actions -- Application to crowds -- CAROSA system overview -- PAR system -- Actionary -- Agent process -- Processing the four action types -- Initializing a scenario -- Building modeling -- Cell and portal graph automatic generation -- Generate cell and portal graph for each floor -- Identify stairs and link floors through new cells -- Identify and store walls -- Identify and store obstacles -- Precalculating data for real-time simulation -- Layout of environment -- Character profiles -- Creating groups -- Constructing actions -- Refining the simulation -- Effects of changes to the environment -- Modifying roles -- Scripting characters -- Evaluating crowds -- Feature comparison -- Low-level features -- Middle-level features -- High-level features -- Summary -- Comparison to real-world data -- Sensor data -- Action statistics -- Validation through the society of fire protection engineers guide -- User evaluations -- Presence in virtual worlds -- Important egocentric features -- Experimental evidence from the literature -- Pilot experiment -- Initial results and future work -- Conclusions on presence as a validation method -- Summary.

Abstract freely available; full-text restricted to subscribers or individual document purchasers.

Compendex

INSPEC

Google scholar

Google book search

There are many applications of computer animation and simulation where it is necessary to model virtual crowds of autonomous agents. Some of these applications include site planning, education, entertainment, training, and human factors analysis for building evacuation. Other applications include simulations of scenarios where masses of people gather, flow, and disperse, such as transportation centers, sporting events, and concerts. Most crowd simulations include only basic locomotive behaviors possibly coupled with a few stochastic actions. Our goal in this survey is to establish a baseline of techniques and requirements for simulating large-scale virtual human populations. Sometimes, these populations might be mutually engaged in a common activity such as evacuation from a building or area; other times they may be going about their individual and personal agenda of work, play, leisure, travel, or spectator. Computational methods to model one set of requirements may not mesh well with good approaches to another. By including both crowd and individual goals and constraints into a comprehensive computational model, we expect to simulate the visual texture and contextual behaviors of groups of seemingly sentient beings.

Also available in print.

Title from PDF t.p. (viewed on December 3, 2008).