Open Source for Handling IndoorGML
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 183 | |
| Author | ||
| License | CC Attribution - NonCommercial - ShareAlike 3.0 Germany: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this | |
| Identifiers | 10.5446/32133 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language | ||
| Producer | ||
| Production Year | 2015 | |
| Production Place | Seoul, South Korea |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
00:00
Dual spaceEndliche ModelltheorieSpacetimeGroup actionRepresentation (politics)SoftwareGraph (mathematics)Prime numberSemantics (computer science)Object (grammar)CausalityDimensional analysisCartesian coordinate systemDuality (mathematics)Three-dimensional spaceBoundary value problemCellular automatonModule (mathematics)Core dumpPlastikkarteOpen sourceExtension (kinesiology)Different (Kate Ryan album)Term (mathematics)Slide ruleFigurate numberConnected spaceGeometryLatent heatNetwork topologyComputer configuration2 (number)Web pageCorrespondence (mathematics)Flow separationFunctional (mathematics)Attribute grammarInformationLocation-based serviceService (economics)Game controllerData modelConstructor (object-oriented programming)Presentation of a groupCombinational logicAdditionScripting languageImplementationForestAreaCarry (arithmetic)View (database)outputInternetworkingMedical imagingDisassemblerMetropolitan area networkDampingAuthorizationNatural numberSource codeWindowCASE <Informatik>SurgeryOcean current1 (number)Arithmetic meanElectric generatorPhysical lawPrisoner's dilemmaBoss CorporationWater vaporWhiteboardPatch (Unix)Thresholding (image processing)State of matterFisher's exact testNeuroinformatikElement (mathematics)Software testingUniverse (mathematics)Special unitary groupAxiom of choiceData managementPressureComputer animation
08:39
Arithmetic progressionQuicksortInformationHybrid computerEndliche ModelltheorieMathematicsExtension (kinesiology)Process (computing)PiFunctional (mathematics)Semiconductor memoryElectric generatorEngineering physicsLocal ringFunctional (mathematics)AreaSlide rulePower (physics)MereologyData managementRepresentation (politics)Presentation of a groupConsistencyMedical imagingShooting methodSocial classMachine visionLevel of measurementDataflowRotationPhysical lawSubsetNumbering schemeBarrelled spaceGodForm (programming)Marginal distributionRange (statistics)SoftwareInclusion mapObject (grammar)Focus (optics)1 (number)String (computer science)ForestComputer programBitGeometryNetwork topologyOrder (biology)Web pageCartesian coordinate systemOpen sourceCompilerBasis <Mathematik>Decision theoryVector potentialLevel (video gaming)Server (computing)Core dumpTheoryDemo (music)PlanningKeyboard shortcutJava appletPlug-in (computing)Projective planeData storage deviceComputer architectureSpacetimeCollision detectionMeasurementData modelUtility softwareComplex (psychology)Module (mathematics)Asynchronous Transfer ModeFile viewerStack (abstract data type)Traffic reportingSoftware developerComputer animation
17:14
Level (video gaming)BitWeb crawlerConnected spaceSpacetimeRight angleGodSoftwareCASE <Informatik>MetadataSystem callState of matterEndliche ModelltheorieMoment (mathematics)Rule of inferenceElement (mathematics)RankingBoundary value problemCore dumpRepresentation (politics)Different (Kate Ryan album)Planar graphComputer animation
20:54
Level (video gaming)Computer animation
Transcript: English(auto-generated)
00:03
Hello everyone, I'm Dongwook So from Busan National University. The topic of this presentation is open source for handling indoor GML. In this presentation, we will learn about what is the indoor GML and what is required tools for handling indoor GML.
00:27
The presentation follows this outline. First, I will introduce background and basic concept of indoor GML and talk about open source for handling indoor GML.
00:43
There are several standards for indoor space. City GML is one of them. City GML focuses on representation, indoor objects such as rooms, doors, and walls. So, city GML is not proper to represent indoor network topology.
01:02
For example, in city GML, there is a different problem to represent the path from R1 to R3. Network topology can be derived from geometric object in city GML, but that cause much computation.
01:20
So, a data model to represent such as network topology was required. Indoor GML can be a solution to this problem. Indoor GML has been recently published as OGC Centers, and it is an application schema
01:41
of GML based on cellular space model and can represent indoor space and set of cells with geometric, topological, and semantic attributes. Indoor GML can represent connectivity, which is characteristic of indoor space.
02:00
It consists of core module and navigation module as extension module. Core module include the page concept of indoor GML, and you can extend core module for specific purpose. Navigation module is one of extension module as navigation purpose. I'm going to explain features of indoor GML,
02:24
geometric, topological, and semantic representation. This slide shows geometric representation in indoor GML. There are three option to represent geometry of each cell. Option one is to use external reference
02:42
to external resources. If indoor spatial data in such as city GML exist, you can just link to corresponding elements. Option two is to represent geometry in indoor GML document. Option three is no geometry because a cell
03:03
can be represented without geometry. We don't need to construct indoor GML data from the script if there are indoor data in other data format. We can combine data in other formats
03:20
and network data in indoor GML and can use it for application. For example, there are several application for service for handicapped persons, emergency control service, and indoor LBS. According to Poincare duality,
03:41
3D space in primal space correspond to dual dimension node in dual space. Similarly, 2D space boundary correspond to one dimension edge in dual space. In indoor GML, nodes and edge are called set and transition respectively.
04:01
3D and network graph about indoor space consist of node and edge. There are two different model to represent wall and doors. The first, in thin wall model, it assume that a wall or door has no thickness.
04:26
This slide shows an example of thin model in topographic space and dual space. There are three cell spaces considering exterior, R1 and R2. R1 is surrounded with B1, B2, and D1.
04:42
In dual space, R1 is a node, and B1, B2, D1 are represented as edge. We can walk through D1, so D1 is navigable edge, but we can't pass through B2, so B2 is non-nevigable edge.
05:03
The second is thick wall model. It assume that wall or door has thickness. In this case, a wall or door considered as a cell. So doors and walls are represented as nodes in dual space.
05:21
There are two types of relationship between nodes, connectivity and adjacency. Indoor space can be interpreted into different purposes. This figure shows an example of representing indoor space
05:40
in different manners in terms of purpose. We assume that there is a step in Room 3. We can walk through the step, but we can't draw a card through the step.
06:01
So Room 3 can be divided Room 3A and Room 3B. And sensor space, such as Wi-Fi coverage, can be represented as a layer, as well as topographic space.
06:24
You can use just one layer altogether, depending on purpose, and we call this concept multi-layered space model. Multi-layered space model can be represented as multi-layered graph.
06:42
In multi-layered graph, a layer is represented as a graph and topology between layers is represented as interlayer connection. For example, Room 3, the node of Room 3 corresponds to nodes of Room 3A and Room 3B,
07:03
so they are connected by interlayer connection. The types of topology in interlayer connection are equals, contains, within, crosses, and intersects. So far, we learned this concept of indoor GMM.
07:21
From now, we talk about the open source software for handling indoor GMM. As indoor GMM has been recently published, so there are a few tools dealing with it. Open source tool for handling indoor,
07:41
open source tool for handling spatial data has been developed, but they are not specialized in indoor space. So, we should develop new tools or implement additional functions for indoor space on existing tools. The first, because data is essential,
08:03
so we need tools for generating and editing the data. JOSM is one of the popular tools that generating and editing geospatial data in outdoor space.
08:20
For indoor space, data editing tools should be able to edit indoor spatial and network information. The second, we need data format handler and the indoor GMM and other data format. For indoor GMM, data format handler
08:41
should support representative data models such as indoor GMM, GV-GMM, and IFC. Third, spatial DBMS that store not only geometric objects,
09:02
but also network data is needed. Similar to PostGIS, a data model for network data can be extended from Postgres SQL. The fourth, a data server that can perform
09:20
very securely like JOSM server is needed for indoor geospatial data. Finally, we need a viewer for indoor 2D or 3D space. For 2D indoor viewer, UI that switch or flow in multi-story building
09:42
is necessary. And for 3D, the viewer should be support various navigation modes and it is recommended that VHX engine including collision detection
10:03
is used in the 3D indoor viewer. Now, our report of tools has been developed or our plan about these tools.
10:23
I located a project for indoor asset management. There are many projects about indoor outdoor areas. Among them, the project of application for editing indoor GMM data is in progress.
10:41
This is the application for editing indoor GMM data in iLocate project. Now, it supports importing, exporting indoor GMM data and it will support the function that editing indoor network data.
11:02
We have two plan of our data format handler. The first is adding indoor GMM model to GDAR or GR. The second is we will develop data format combating tool. It provides the combating function between indoor GMM and other data formats.
11:23
And we will define Java geometry package as geometry classes between data formats. So the tool will support combating using this Java geometry.
11:40
Before combating indoor GMM to other data formats, we should be able to supervise data in memory. So we generated Java package for indoor GMM using Jaxby. Jaxby is Java API that represents the Java class to XML. Jaxby provides two main features.
12:02
The ability to measure Java object into XML and measure XML back into Java object. And Java classes can extract it from XML schema by Jaxby binding compiler.
12:20
Similarly, we extracted Java theory general package from XML schema. So we can make indoor GMM and theory general data into Java object. So can use it for data format combating tool.
12:43
Last, our plan of our data server is ISA server. ISA server is handled in previous presentation, so talk about briefly. This slide shows the architecture of ISA server. ISA server consists of GeoServer
13:01
and GeoTool's core library. GeoTool's extensions and GeoTool's plugins. We'll focus on parts that we are involved in, extension and plugins. ISA engine improves 3D geometry and complex feature. But these two modules can center around as GeoTool's extensions.
13:25
In conclusion, in order to utilize indoor GMM in various fields, open source tools that provide a basis of developing applications are required. The types of tools range from editing tool to viewer.
13:44
Editing tool, their format handler, their server are in the middle of being developed. Thank you for your attention. Any questions?
14:01
We have about five minutes for Q and A. Is there any public demo of the iLocate portal? I have been trying to have a look at it a few weeks ago and I haven't been able to see a functioning web page
14:21
where I can just see indoor GMM data with all the stack. Is that available somewhere or is there a timeline to making such a public demo?
14:46
It's not open source-y, you know. We're going to do these tools, but we're not going to show them and we're going to go to a Phosphor-G because, you know, show me your tools, show me the working place. Please, please, please, please.
15:19
Okay, please make an announcement when possible, okay.
15:33
City indoor GMM. Is there a relationship between indoor GMM and BIM? And is there a way to convert the current BIM
15:42
to indoor GMM? When, you know, designed indoor GMM, we considered the city GMM models, not only the city GMM but also the PIM IFC model.
16:03
So this, I mean, indoor GMM does not deal with the geometry part. I mean that, for example, this room consists by, you know, just kind of this surly, right? But we are focused on network data, right?
16:22
To, you know, for example, routing data for navigation. So, yes, maybe we can extend. I mean, if you have data, PM data, and you want to make a network, we can automatically or manually
16:42
can build that indoor GMM data and we can combine. So that's why you can see the slide, the indoor GMM plus and the city GMM, maybe KEMM or just flow plan image. Yeah, you can join to, you know, develop any application, indoor application, yeah.
17:02
So in short is that if we can create a simplified BIM, BIM, BIM, and in short create a simple one, and then we could actually very easily change the, because it's all defined in the BIM, right? All the walls and everything. Yeah, so if you create a simplified thing,
17:21
we could actually create a indoor GMM very quickly. Right? Thank you. Remember, already we built some, yeah, indoor tool, indoor GMM tool.
17:42
It's just manually, we can, if we have a background image, just pick a node and add, yeah, like that.
18:18
So in indoor GMM, that module,
18:24
a core module, they define space and space boundary. So if BIM have a space, yeah, it's just, we can link, we call the node state. So state and it's a relate,
18:41
you know, we just link the, what is it? IFC space, right? You know, the element name, maybe the IFC space. I'm not sure. But yeah, you can link it. It means you already know the relationship
19:03
between the node and the space, IFC space. So it can, you know, solve. But we don't, I don't know about the metadata at the moment that between, you know, yeah, different format.
19:23
Any more questions? Okay, I have one, just a short question. Is this indoor GMM, is it possible to represent network between flaws?
19:48
Yes. Yeah. No, what I mean is between different layers, not on the same planar.
20:00
Yeah, it's a different layer. So different levels, what I mean is a different level from level one to level two access. Will it be possible? Florian. If, you know, there are two space
20:21
and they share the boundary, it can be a door or it can be a wall, right? So in this case, this room, maybe next room have separate by wall. Or that in this case, we have, it's a adjacency relationship,
20:42
and this room and corridor has a connection, right? So different relationships. So as you mentioned, maybe we can, if there is some stair, then it's connectivity. We can represent it connectivity, but if the lower, you know, level and upper level
21:02
can just, well, we're gentle of the agency like that. Okay.