supporting spatial planning with …2 o 1 meet o 2 o 1 overlap o 2 o 1 covers o 2 o 1 coveredby o 2...
Embed Size (px)
TRANSCRIPT
-
W NEI
TECHNISCHEUNIVERSITÄTWIENVienna University of Technology
DEPARTMENT FORGEODESY ANDGEOINFORMATION
Ge inf
RESEARCH GROUPGEOINFORMATION
SUPPORTING SPATIAL PLANNING WITH
QUALITATIVE CONFIGURATION ANALYSIS
20.05.2013CORP 2013
Paolo Fogliaroni and Gerhard Navratil
venerdì 17 gennaio 14
-
Ge infW NEISPATIAL PLANNING TODAY SNAPSHOT
• Spatial planning strongly relies on Spatial Information Systems (SIS):
• Computer-Aided Design
• Geographic Information Systems
• SIS offer continuously more powerful spatial data analysis and management options
• Yet, SIS largely lack support for natural human-computer interaction
venerdì 17 gennaio 14
-
Ge infW NEI
SIS VS HUMANS
• SIS → quantitative representations (raster/vector) + numerical operationsE.g.: The angular lake-park distance is 158.9°
• Humans → qualitative representations + reasoning E.g.: The lake is to the west of the park
• Today, translation efforts to map qualitative spatial representation into a quantitative (numerical/geometric) one is up to SIS user
158.9°Lake Park
venerdì 17 gennaio 14
-
Ge infW NEI
QUALITATIVE INFORMATION
• Qualitative information is, by its own nature, somewhat “vague” and “imprecise”
• It is not a substitute of quantitative information
• Yet, when it comes to interaction with human beings, it plays a fundamental role
• How to embody qualitative models into SIS?
venerdì 17 gennaio 14
-
Ge infW NEIQUALITATIVE SPATIAL
REPRESENTATION AND REASONING• Qualitative Spatial Representation and Reasoning (QSR) is a subfield of Artificial Intelligence providing
so-called qualitative spatial calculi
• A qualitative calculus
• is cognitive suitable
• focuses on a single aspect of space (e.g. topology, direction, or distance)
• consists of two main items:
• a finite set of symbols called qualitative spatial relations: used for representational purposes
• a finite set of inference rules that allow for symbolic reasoning
o1 Disjoint o2
o1 Meet o2
o1 Overlap o2
o1 Covers o2
o1CoveredBy o2
o1 Equal o2
o1 Contains o2
o1 Inside o2 The 9-intersection model defines the 8 topological relations that can hold
between two spatial objects;on the left depicted and
arranged according to their conceptual neighborhood.
Egenhofer, M.J. (1989).venerdì 17 gennaio 14
-
Ge infW NEIQUALITATIVE SPATIAL
INFORMATION SYSTEM
Query/Analysis/Editing
Spatial Information System
Geometry
Raster/Vector
Low-level operations
Quantitative Data storage
Spatial DB
Web/Local visualization
Fogliaroni, P. (2012).venerdì 17 gennaio 14
-
Ge infW NEIQUALITATIVE SPATIAL
INFORMATION SYSTEM
Query/Analysis/Editing
Spatial Information System
Geometry
Raster/Vector
Low-level operations
Quantitative Data storage
Spatial DB
Pool of Qualitative
Calculi
+ Qualitative ExtensionWeb/Local visualization
Fogliaroni, P. (2012).venerdì 17 gennaio 14
-
Ge infW NEIQUALITATIVE SPATIAL
INFORMATION SYSTEM
Query/Analysis/Editing
Spatial Information System
Geometry
Raster/Vector
TopologyDistanceDirectionVisibility
⋮Low-level operations
Quantitative Data storage
Spatial DB
Pool of Qualitative
Calculi
Qual.Rels
+ Qualitative ExtensionWeb/Local visualization
Fogliaroni, P. (2012).venerdì 17 gennaio 14
-
Ge infW NEIQUALITATIVE SPATIAL
INFORMATION SYSTEM
Query/Analysis/Editing
Spatial Information System
Geometry
Raster/Vector
TopologyDistanceDirectionVisibility
⋮Low-level operations
Quantitative Data storage
Spatial DB
Pool of Qualitative
Calculi
Qualitative Data storage
Qualitative Representation
Qual.Rels
+ Qualitative Extension
Qual.Repr.
Web/Local visualization
Fogliaroni, P. (2012).venerdì 17 gennaio 14
-
Ge infW NEIQUALITATIVE SPATIAL
INFORMATION SYSTEM
Query/Analysis/Editing
Spatial Information System
Geometry
Raster/Vector
TopologyDistanceDirectionVisibility
⋮Low-level operations
Quantitative Data storage
Spatial DB
Pool of Qualitative
Calculi
Qualitative Data storage
Qualitative Representation
Qual.Rels
+ Qualitative Extension
Qual.Repr.
Web/Local visualization
Qualitative query/
Reasoning
Fogliaroni, P. (2012).venerdì 17 gennaio 14
-
Ge infW NEIQUALITATIVE
REPRESENTATION
o1
o2
o3
Quantitative representationVector
o1
o3
o2
Disjoint,West,...
Disjoint,East,...
Disj
oint
,East,
...
Disj
oint
,Wes
t,...
Meet,Ea
st
Meet,W
est
Qualitative representationQualitative Constraint Network (QCN)
venerdì 17 gennaio 14
-
Ge infW NEIQUALITATIVE
REPRESENTATION
o1
o3
o2
Disjoint,West,...
Disjoint,East,...
Disj
oint
,East,
...
Disj
oint
,Wes
t,...
Meet,Ea
st
Meet,W
est
Qualitative representationQualitative Constraint Network (QCN)
Qualitative QueriesNatural language sentence
I’m looking for a pair of objects such that they are in touch and one lies west of the other!
Does object 1 touch object 2?
o1 o2
or a sketch
venerdì 17 gennaio 14
-
Ge infW NEI
SPATIAL PLANNING SUPPORT• Qualitative Spatial Information System +
• spatial object ontology: e.g. lake, park, building, house...
• spatial plan ontology: e.g. public green, residential area...
• Each designed plan is stored in the database together with the corresponding qualitative config
Lake(o1)
Park (o2)
Road (o3)
Geometric Plan Design Qualitative Configuration
Lake(o1)
Road(o3)
Park(o2)
Disjoint,West,...
Disjoint,East,...
Disj
oint
,East,
...
Disj
oint
,Wes
t,...
Meet,Ea
st
Meet,W
est
venerdì 17 gennaio 14
-
Ge infW NEI
SPATIAL PLANNING SUPPORT• Qualitative Spatial Information System +
• spatial object ontology: e.g. lake, park, building, house...
• spatial plan ontology: e.g. public green, residential area...
• Each designed plan is stored in the database together with the corresponding qualitative config
Lake(o1)
Park (o2)
Road (o3)
Geometric Plan Design Qualitative ConfigurationLake(o1)
Road(o3)
Park(o2)
Disjoint,West,...
Disjoint,East,...
Disj
oint
,East,
...
Disj
oint
,Wes
t,...
Meet,Ea
st
Meet,W
est
venerdì 17 gennaio 14
-
Ge infW NEI
SPATIAL PLANNING SUPPORT• Qualitative Spatial Information System +
• spatial object ontology: e.g. lake, park, building, house...
• spatial plan ontology: e.g. public green, residential area...
• Each designed plan is stored in the database together with the corresponding qualitative config
Lake(o1)
Park (o2)
Road (o3)
Geometric Plan Design Qualitative ConfigurationLake(o1)
Road(o3)
Park(o2)
Disjoint,West,...
Disjoint,East,...
Disj
oint
,East,
...
Disj
oint
,Wes
t,...
Meet,Ea
st
Meet,W
est
Park (o2)
Lake(o1)
Road (o3) Lake(o1)
Road(o3)
Park(o2)
Contains
Inside
Disj
oint
Disj
oint
Meet
Meet
venerdì 17 gennaio 14
-
Ge infW NEI
SPATIAL PLANNING SUPPORT• Qualitative Spatial Information System +
• spatial object ontology: e.g. lake, park, building, house...
• spatial plan ontology: e.g. public green, residential area...
• Each designed plan is stored in the database together with the corresponding qualitative config
Lake(o1)
Park (o2)
Road (o3)
Geometric Plan Design Qualitative ConfigurationLake(o1)
Road(o3)
Park(o2)
Disjoint,West,...
Disjoint,East,...
Disj
oint
,East,
...
Disj
oint
,Wes
t,...
Meet,Ea
st
Meet,W
est
Park (o2)
Lake(o1)
Road (o3)Lake(o1)
Road(o3)
Park(o2)
Contains
Inside
Disj
oint
Disj
oint
Meet
Meet
venerdì 17 gennaio 14
-
Ge infW NEI
SPATIAL PLANNING SUPPORT• Qualitative Spatial Information System +
• spatial object ontology: e.g. lake, park, building, house...
• spatial plan ontology: e.g. public green, residential area...
• Each designed plan is stored in the database together with the corresponding qualitative config
Lake(o1)
Park (o2)
Road (o3)
Geometric Plan Design Qualitative ConfigurationLake(o1)
Road(o3)
Park(o2)
Disjoint,West,...
Disjoint,East,...
Disj
oint
,East,
...
Disj
oint
,Wes
t,...
Meet,Ea
st
Meet,W
est
Park (o2)
Lake(o1)
Road (o3)Lake(o1)
Road(o3)
Park(o2)
Contains
Inside
Disj
oint
Disj
oint
Meet
Meet
Park (o2)
Lake(o1)Road (o3)
Lake(o1)
Road(o3)
Park(o2)
Covers
Covered By
Disj
oint
Disj
oint
Meet
Meet
venerdì 17 gennaio 14
-
Ge infW NEI
SPATIAL PLANNING SUPPORT• Qualitative Spatial Information System +
• spatial object ontology: e.g. lake, park, building, house...
• spatial plan ontology: e.g. public green, residential area...
• Each designed plan is stored in the database together with the corresponding qualitative config
Lake(o1)
Park (o2)
Road (o3)
Geometric Plan Design Qualitative ConfigurationLake(o1)
Road(o3)
Park(o2)
Disjoint,West,...
Disjoint,East,...
Disj
oint
,East,
...
Disj
oint
,Wes
t,...
Meet,Ea
st
Meet,W
est
Park (o2)
Lake(o1)
Road (o3)Lake(o1)
Road(o3)
Park(o2)
Contains
Inside
Disj
oint
Disj
oint
Meet
Meet
Park (o2)
Lake(o1)Road (o3)
Lake(o1)
Road(o3)
Park(o2)
Covers
Covered By
Disj
oint
Disj
oint
Meet
Meet
venerdì 17 gennaio 14
-
Ge infW NEICOMPUTING PROTOTYPICAL
PLANS1.Normalize qualitative configs by
1.1.computing maximum common set of spatial objects
1.2.removing non-‐common objects (and relations they are involved in)
2.Compare normalized configs via graph matching techniques to find the minimum common subgraph
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
Lake
venerdì 17 gennaio 14
-
Ge infW NEICOMPUTING PROTOTYPICAL
PLANS1.Normalize qualitative configs by
1.1.computing maximum common set of spatial objects
1.2.removing non-‐common objects (and relations they are involved in)
2.Compare normalized configs via graph matching techniques to find the minimum common subgraph
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
Lake
venerdì 17 gennaio 14
-
Ge infW NEICOMPUTING PROTOTYPICAL
PLANS1.Normalize qualitative configs by
1.1.computing maximum common set of spatial objects
1.2.removing non-‐common objects (and relations they are involved in)
2.Compare normalized configs via graph matching techniques to find the minimum common subgraph
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
Lake
venerdì 17 gennaio 14
-
Ge infW NEICOMPUTING PROTOTYPICAL
PLANS1.Normalize qualitative configs by
1.1.computing maximum common set of spatial objects
1.2.removing non-‐common objects (and relations they are involved in)
2.Compare normalized configs via graph matching techniques to find the minimum common subgraph
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
venerdì 17 gennaio 14
-
Ge infW NEISUPPORTING SPATIAL
PLANNING
• Our system supports spatial planners in 3 different stages:
• Preliminary phase: site localization
• Plan setup: kickstart configuration
• Plan development: design assistance
venerdì 17 gennaio 14
-
Ge infW NEI
LOCALIZING A PLAN SITE
• Finding a suitable location for a new plan consists in finding land parcels satisfying a number of constraints
• Part of such constraints are purely spatial: search for a spatial configuration of objects arranged in a certain manner
• Strict constraints are better expressed with mathematical equations
• Loose constraints are more easily expressed in natural language without resorting to complex equations, inequalities and conditions
• Our system allows for complementing standard search methods via hybrid quantitative-qualitative spatial queries
venerdì 17 gennaio 14
-
Ge infW NEI
DESIGN KICKSTARTER
SpatialDB
venerdì 17 gennaio 14
-
Ge infW NEI
DESIGN KICKSTARTERPlan category: public green
Prototypical configuration
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
SpatialDB
venerdì 17 gennaio 14
-
Ge infW NEI
DESIGN KICKSTARTERMain Road
Park Running Track
Parking Lot
Meet
Meet
Contains
venerdì 17 gennaio 14
-
Ge infW NEI
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
DESIGN ASSISTANTMain Road
Park Running Track
Parking Lot
Meet
Meet
Contains
Prototypicalconfiguration
Plan design
venerdì 17 gennaio 14
-
Ge infW NEI
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
DESIGN ASSISTANT
Main Road
Parking Lot
Meet
Meet
Prototypicalconfiguration
Plan design
Currentconfiguration
Contains Park Running Track
venerdì 17 gennaio 14
-
Ge infW NEI
Main Road
Park Running Track
Parking Lot
Meet
Meet
Contains
DESIGN ASSISTANT
Main Road
Parking Lot
Meet
Meet
Prototypicalconfiguration
Plan design
Currentconfiguration
DisjointPark Running Track
venerdì 17 gennaio 14
-
Ge infW NEIFITTING PLANS TO PUBLIC
EXPECTATIONS• Public Participatory Geographic Information Systems (PPGIS)
are web platforms designed to collect public opinion of a topic of interest
• PPGIS allows for adapting prototypical plans to public expectation in two ways:
• ratings collection
• public envisioning
venerdì 17 gennaio 14
-
Ge infW NEI
RATING COLLECTION
• Spatial plan will eventually affect real world
• The resulting environment is best assessed by its users
• PPGIS can be used to collect user feedbacks and generate an overall rating of a certain environment
• The rating is associated to the corresponding plan design in the database
• Ratings are used to weight plans: high-scored designs play an heavier role in the prototypical plan generation process
venerdì 17 gennaio 14
-
Ge infW NEIPROTOTYPICAL PLAN FROM
PEOPLE EXPECTATION• Web design tools can be used to let people sketching the new
environment as they expect it
• Qualitative configurations can be obtained from such sketches and associated to a certain plan category
B3 (www.geogameslab.com) is a web platform that allows laypeople to draw an environment via drag and drop
Poplin, A. (2012).venerdì 17 gennaio 14
http://www.geogameslab.comhttp://www.geogameslab.com
-
Ge infW NEI
THANK YOU FOR YOUR ATTENTION!
QUESTIONS ARE WELCOME
Egenhofer, M.J. (1989). A formal definition of binary topological relationships. In: Litwin, W., Schek, H.J. (eds.), FODO 1989, 3rd International Conference on Foundations of Data Organization and Algorithms, Lecture Notes in Computer Science, vol. 367, pp. 457–472. Springer-Verlag
Fogliaroni, P. (2012). Qualitative Spatial Configuration Queries – Towards Next Generation Access Methods for GIS. Ph.D. Thesis (http://nbn-resolving.de/urn:nbn:de:gbv:46-00102731-11)
Poplin, A. (2012). Playful public participation in urban planning: A case study for online serious games. Computers, Environment and Urban Systems.
venerdì 17 gennaio 14
http://nbn-resolving.de/urn:nbn:de:gbv:46-00102731-11http://nbn-resolving.de/urn:nbn:de:gbv:46-00102731-11