D. Moser, A. Riener, K. Zia, A. Ferscha D Department for Pervasive Computing, JKU Linz/Austria t t f P i C ti JKU Li /A t i “Parallel Simulation of Social Agents using Cilk and OpenCL” DS-RT 2011 15th International Symposium on Distributed Simulation and Real Time Applications September 4,-7, 2011, Salford/Manchester, UK Dr. Andreas Riener JKU Linz, Department for Pervasive Computing This work is supported under the FP7 ICT FET Altenberger Straße 69, A-4040 Linz program of the European Commission under grant program of the European Commission under grant www.pervasive.jku.at/about_us/staff/riener agreement No 231288 (SOCIONICAL)
Socio Technical Systems: Motivation for Modeling/Simulation Socio-Technical Systems: Motivation for Modeling/Simulation “Development of complexity science based modeling, prediction and simulation methods Development of complexity science based modeling, prediction and simulation methods for large scale socio-technical systems in an AmI based smart environment” � Experiments : standard way of collecting evidence in such (dynamic) systems � allow to analyze situations, person behavior and to interview test participants � Experimentation is, however, not possible in large social systems � undesirable/unaccepted � repeatability not given � dangerous (for involved persons, infrastructure) � impossible (in terms of scale or behavior) � impossible (in terms of scale or behavior) (i) evacuation of a large megacity with million of peoples is not possible (ii) different behavior of entities/persons on artificial/simulated hazards compared to a real incident; generating of a “real nuclear incident” is not feasible � Solution : Simulated interaction of agents based on realistic behavioral rules � “agent” = entity with realistic behavior and interaction capabilities � AmI technology to “enhance” agents (FOV knowledge etc ) AmI technology to enhance agents (FOV, knowledge, etc.) � real underlying space model DS-RT 2011 // September 4-7, 2011 // Manchester, UK // Slide 2 A. Riener, JKU Linz
Socio Technical Systems: Motivation for Modeling/Simulation (2) Socio-Technical Systems: Motivation for Modeling/Simulation (2) Agent-based modeling (ABM) Agent based modeling (ABM) � A widely used analytical method capable to represent individual entities and their interactions [Gilbert2008] � Resource intensive – using a single machine, simulation of only small g g y models/local behavior possible � Only suited for small- to medium-sized problems (single workstation) [Zia2010] Discrepancy “resources” ↔ “large scale”? � Due to advancements in processing power (GPGPU) and/or cluster technology (PDS) no longer a problem… Close-to-reality results � Developments in cognitive social modeling allows for the first time for close-to- reality simulation of social or collective phenomena (e.g., group formation) lit i l ti f i l ll ti h ( f ti ) � Further model up-scales allows ABM to explain the emergence of higher order patterns (movement dynamics in traffic jams, behavioral patterns in global social networks, social segregation across populations) networks, social segregation across populations) DS-RT 2011 // September 4-7, 2011 // Manchester, UK // Slide 3 A. Riener, JKU Linz
Socio Technical/Social Systems: Agent Based Modeling Socio-Technical/Social Systems: Agent-Based Modeling Designing agents in software Designing agents in software for each agent a in simulation for each agent a in simulation stats = synchronize(a); ( individual representation of an agent ) if AmI-assisted agent update-intentions else update-proximity-parameters � Perception : Agents can perceive their neighborhood, i.e. they can determine for each exit e what agents (including space agents) are dist = decision-param (e); distance/belief etc. hope a (e)= … ; based on group emotions, individualism in their vicinity fear a (e)= … ; based on group emotions, individualism attract a (e)= …; based on group emotions, individual… a � Performance : How do they perform their curr-exit = choose exit with max attract activity which may include motion, curr-dir = get-direction curr-exit ; floor field communication (interaction) and action heading = curr-dir ; setting heading of the agent ( (changing states of itself or other agents) g g g ) MOVE update-proximity-parameters � Memory : They have a memory where For each agent n in neighborhood of a for each exit e they can record their action and states update group fear, hope and attract which may include the history which may include the history update-intentions � update-proximity-parameters Policy : they have a set of rules, heuristics, For each agent n in the neighborhood of a or strategies that determines, given their belief n (curr-exit)= … ; based on belief/trust of group present situation and their history, what present situation and their history what trust (a)= trust n (a)= … ; based on belief of a on curr-exit ; b d b li f f it behaviors they would now carry out DS-RT 2011 // September 4-7, 2011 // Manchester, UK // Slide 4 A. Riener, JKU Linz
Socio Technical/Social Systems: Agent Based Modeling Socio-Technical/Social Systems: Agent-Based Modeling Designing agents in software Designing agents in software for each agent a in simulation for each agent a in simulation stats = synchronize(a); � Perception : Agents can perceive their if AmI-assisted agent update-intentions neighborhood, i.e. they can determine else update-proximity-parameters what agents (including space agents) are h t t (i l di t ) for each exit e in their vicinity dist = decision-param (e); distance/belief etc. hope a (e)= … ; based on group emotions, individualism fear a (e)= … ; based on group emotions, individualism � Performance : How do they perform their attract a (e)= …; based on group emotions, individual… a activity which may include motion, communication (interaction) and action curr-exit = choose exit with max attract curr-dir = get-direction curr-exit ; floor field (changing states of itself or other agents) heading = curr-dir ; setting heading of the agent MOVE � Memory : They have a memory where update-proximity-parameters they can record their action and states For each agent n in neighborhood of a which may include the history for each exit e update group fear, hope and attract � Policy : they have a set of rules, heuristics, update-intentions or strategies that determines, given their update-proximity-parameters For each agent n in the neighborhood of a present situation and their history, what belief n (curr-exit)= … ; based on belief/trust of group behaviors they would now carry out behaviors they would now carry out trust n (a)= … ; based on belief of a on curr-exit trust (a)= ; b d b li f f it DS-RT 2011 // September 4-7, 2011 // Manchester, UK // Slide 5 A. Riener, JKU Linz
Socio Technical/Social Systems: Agent Based Modeling Socio-Technical/Social Systems: Agent-Based Modeling Designing agents in software Designing agents in software for each agent a in simulation for each agent a in simulation stats = synchronize(a); � Perception : Agents can perceive their if AmI-assisted agent update-intentions neighborhood, i.e. they can determine else update-proximity-parameters what agents (including space agents) are h t t (i l di t ) for each exit e in their vicinity dist = decision-param (e); distance/belief etc. hope a (e)= … ; based on group emotions, individualism fear a (e)= … ; based on group emotions, individualism � Performance : How do they perform their attract a (e)= …; based on group emotions, individual… a activity which may include motion, communication (interaction) and action curr-exit = choose exit with max attract curr-dir = get-direction curr-exit ; floor field (changing states of itself or other agents) heading = curr-dir ; setting heading of the agent MOVE � Memory : They have a memory where update-proximity-parameters they can record their action and states For each agent n in neighborhood of a which may include the history for each exit e update group fear, hope and attract � Policy : they have a set of rules, heuristics, update-intentions or strategies that determines, given their update-proximity-parameters For each agent n in the neighborhood of a present situation and their history, what belief n (curr-exit)= … ; based on belief/trust of group behaviors they would now carry out behaviors they would now carry out trust n (a)= … ; based on belief of a on curr-exit trust (a)= ; b d b li f f it DS-RT 2011 // September 4-7, 2011 // Manchester, UK // Slide 6 A. Riener, JKU Linz
Recommend
More recommend
