Transcript Circulation Simulation

Circulation Simulation
Andrew Moeding
Simulation Types
• Traffic flow pattern simulation
• Building/pedestrian circulation simulation
Who is using it ?
• Engineers, Architects, Urban Planners, Corporations,
Universities, Event Organizers, Police
–
–
–
–
Security / egress testing
Walkability studies
Building circulation
Pedestrian / traffic interaction
• Highway / Transportation departments
–
–
–
–
Interchange Design
Signal coordination
Rapid transit operations
Transit station design
– http://horstmann.com/applets/RoadApplet/RoadApplet.html
– http://www.phy.ntnu.edu.tw/oldjava/Others/trafficSimulation/applet.html
– http://www.traffic-simulation.de/
Software Products
•
•
•
•
Trafficware – Sim Traffic
TSI-CORSIM
SimWalk,
Legion Studio, http://www.legion.com/case-studies/sydney-olympicspopup1.php
• AI Implant
• Massive
How it works
• Agent Based Model
–
–
–
–
Agent based models consist of dynamically interacting rule based agents. The systems
within which they interact can therefore create complexity like that which we see in the real
world.
The idea is that a system adapts to internal and external pressures so as to maintain
functionalities. The task of harnessing that complexity requires consideration of the agents
themselves -- their diversity, connectedness, and level of interactions.
The system of interest is simulated by capturing the behavior of individual agents and their
interconnections. Agent-based modeling tools can be used to test how changes in individual
behaviors will affect the overall, emergent system behavior.
http://www.casa.ucl.ac.uk/repast/JohnWard_ThursdayMeeting2006_01_19.pdf
–
http://www.casa.ucl.ac.uk/repast/AndrewCrooks_19_1_06.pdf
– http://www.youtube.com/watch?v=Cw1b_RYi784
–
http://www.youtube.com/watch?v=0pILzhLpMPc&mode=related&search
Wikipedia
How it works
• Multi-Agent System
– (MAS) is a system composed of several software agents, collectively capable of
reaching goals that are difficult to achieve by an individual agent or monolithic
system.
– Multi-agent systems can manifest self-organization and complex behaviors even
when the individual strategies of all their agents are simple.
– The study of Multi-Agent Systems is concerned with the development and
analysis of sophisticated Artificial intelligence problem solving and control
architectures for both single-agent and multiple-agent systems.
– Another paradigm commonly used with MAS systems is the pheromone, where
components "leave" information for other components "next in line" or "in the
vicinity". These "pheromones" may "evaporate" with time, that is their values may
decrease (or increase) with time.
– http://digitalurban.blogspot.com/search/label/3D%20Agents
Wikipedia
How it works
• Multi-Agent System Cont.
– MAS systems are also referred to as "self-organized systems" as they tend to
find the best solution for their problems "without intervention". There is high
similarity here to physical phenomena, such as energy minimizing, where
physical objects tend to reach the lowest energy possible, within the physical
constrained world.
• Crowd Simulation
– The entities - also called agents - are given artificial intelligence, which guides
the entities based on one or more functions, such as sight, hearing, basic
emotion, energy level, aggressiveness level, etc.. The entities are given goals
and then interact with each other as members of a real crowd would. They are
often programmed to respond to changes in environment, enabling them to climb
hills, jump over holes, scale ladders, etc. This system is much more realistic than
particle motion, but is very expensive to program and implement.
Wikipedia
AI Implant
• What it is
– http://www.aiimplant.com/solutions/simulation__training/technical_specifications.htm
• Examples
– http://www.youtube.com/watch?v=lCH0QZFs7iY
– http://www.youtube.com/watch?v=T5VZFxRJ6ss
– http://www.massivesoftware.com/
Sulan Kolatan, Columbia studio
AIA - Architectural Intelligence Agency: In choosing to work with software
specifically created for industrial design and film animation rather than for architectural
design, our studio explicitly engaged the issue of cross-categorical pollination by
problematizing it in the design process itself. In this way, the architectural design process
was affected by a "productive inadequacy". The design tool was not entirely but
somewhat inadequate in that it had not been made to address the conventions of
architectural design but rather those of another kind of design. It was like having to write
with a knife. One had to rethink "writing" through the logic of "cutting" to arrive at
"carving". The studio's intent was to introduce computational methodologies into
architectural design through the use of self-organizing system software. Play was
combined with analytical and speculative thought to diagram and construct architectures
without fixed scale but with set rules. Scalability was to be understood as referring to a
diagram (set of rules) capable of being translated into many scales and -- by extension -contexts. Particularly viable scales and contexts were those where performative affinities
to the diagram already existed. The studio was structured in five phases focusing on
Investigative Play, Identity Programming, Dynamic Chimerization, Variable Inhabitation
and Performative Taxonomy.
Sulan Kolatan, Columbia studio
Sulan Kolatan, Columbia studio
Sulan Kolatan, Columbia studio
Projects
Frank Gesualdi
“We were scripting behavior into
a population of agents and
allowing these agents to mix as
if in a virtual petry dish of sorts.
We were particularly interested
in scripting behavior that was
"programmatic" ie this
population of agents
represented quiet space, this
population was akin to attraction
to groups over 10…”
Additional Research
•
http://www.mukogawa-u.ac.jp/~okazaki/OK/PMOVE/paper1/London.pdf
•
http://www.crowddynamics.com/index.htm
•
http://www.casa.ucl.ac.uk/projects/projectDetail.asp?ID=58
•
http://www.digitalurban.blogspot.com/