D3D-LEM · Dynamic 3D modelling and simulation for landscape and environmental management.
6РП — Действия „Мария Кюри“
- Период
- 2004-09-01 → 2007-08-31
- Финансиране от ЕС
- 504 321 €
- Участници
- 1
- Схема
- EXC
Линиите свързват координатора с партньорите. За проекти отпреди 2014 г. CORDIS не винаги дава точни координати. Тези точки са на ниво град или държава.
Накратко на български
Дигиталните 3D модели и структури от данни помагат за по-точното представяне на пространството, например при автоматичното възстановяване на сгради от лазерни данни. Те подобряват управлението на околната среда, морската навигация и планирането на пътища за евакуация в реално време.
Кратко обяснение, генерирано от езиков модел по текста на CORDIS. Оригиналът е по-долу.
Резултати накратко
Final Activity Report Summary - D3D-LEM (Dynamic 3D modelling and simulation for landscape and environmental management.)
The objectives of the chair were to develop, compile and disseminate a cross-disciplinary set of data structures and operators for working with spatial data. The primary focus was to be on Geographic information systems (GIS), but with strong links to computer science and various applications disciplines. There were four stages: compilation; theoretical development; applications; and training. For all disciplines, the key issue was topology: the way that entities related to each other in space. For computer systems this emphasized data structures and the operations permitting their local update and modification. Structures were initially two-dimensional (2D), with developments leading to three-dimensional (3D) modelling. For automated construction the geometric considerations led to emphasis on the 2D and 3D Voronoi diagram (VD), as well as the Delaunay triangulation due to an increasing emphasis on graph duality as a valuable applications tool. Six PhD students were supervised in the three years of the chair. Their topics included: - landscape visualisation; - kinetic 3D Voronoi diagrams and their application in modelling spatial fields; - automated building reconstruction from LIDAR data; - 3D visualisation for on-ship marine navigation; interactive terrain modelling and applications; and - 3D data structures for real-time building escape-route planning. Theoretical work was concerned with: - the VD and algorithms for interactive modification of the spatial model, in 2D and 3D; - interactive line-segment VD construction for GIS applications and analysis of spatial relationships; - relating CAD and GIS data structures to permit integration of buildings and landscape; oceanographic modelling and interpolation; - quad-edge data structures and CAD Euler operators; and - extending quad edge structures to 3D in order to maintain primal and dual relationships. One co-authored book, 3 book chapters, 8 published journal articles and 42 conference papers have been published to date. A major component of the chair was the presentation of workshops to students of various disciplines. 16 workshops were given, of one to five days in length, at various European universities - some as credit courses. The main theme was the importance of having a clear model of the space one was attempting to model, and on this basis to develop or select the appropriate data structure and algorithms. Hands-on laboratory sessions were available, and also research student seminars to provide discussion and advice on research projects. A major result of the chair was the development of dynamic and kinetic algorithms for Voronoi diagrams in 2D and 3D. Applications were developed for marine navigation, modelling of 2D and 3D fields and interactive cartography. A consistent model has been proposed for many types of spatial and GIS data, integrating a wide variety of data types, topological structures, algorithms and analyses. In particular, the integration of GIS-type 'objects' and 'fields' allows for more consistent data structures and spatial queries. Recent work suggests a hierarchical model for these methods, permitting their extension to large data sets. Automated building reconstruction from laser altimetry data is of considerable interest to many national mapping agencies. Chair research has led to improved methods for building-plane recognition from point-cloud data, as well as flexible algorithms for reconstruction (reverse engineering) of buildings based on the recognition of the topological relationships of these planes, rather than using pre-defined building templates. Significant progress was also made in the development of a true 'volumetric' 3D data structure, based on preserving both primal and dual relationships, that permits true navigation in, and querying of, spatial models containing volumes, faces, edges and vertices - and the attributes associated with these entities. This integrates the GIS concept of spatial entities with identities and attributes, with the CAD model of 3D shells and construction operators. Recent work has been on developing disaster management tools, where linkages between rooms and accessible escape paths are critical, but other applications include geological modelling, etc. Overall, these concepts have been of great utility in the integration of spatial modelling tools between various disciplines. Various researchers are continuing to develop further concepts and applications based on the chair ideas.
Текст от CORDIS, на английски · Данни: CORDIS, © Европейски съюз
Цел на проекта
Geo-informatics is concerned with the collection, display and utilization of spatial information - a key feature of many aspects of planning and decision-making. It is expected that the future of Geo-informatics will heavily involve virtual reality and simulation.The Chair research will attempt to integrate Computer Science advances with the needs of spatial planning. Geographic Information Systems (GIS) were developed for 2D static data display, whereas simulation requires movement, interaction and often the third dimension.This requires major changes in the data structures used to preserve topology (connectedness) between the various components in a changing 3D world. Difficulties occur from the use of an inappropriate perceptual model for some particular problem, and a lack of familiarity of many spatial analysts with the latest Computer Science research.The proposed Chair agenda thus has the following components:- A compilation component, to obtain a sufficient set of spatial structures and operators for manipulating geo-data, based on work by the Chair Holder, Computer Science researchers, and Geo-informatics specialists.- A theoretical component, where gaps in this set of tools would be filled - e.g. by the use of dynamic or kinetic Voronoi diagram s in 2D and 3D; interactive landscape modification for simulation or urban planning; full surface modelling for complex urban landscapes; and dynamic 3D volumetric modelling for geological, marine or atmospheric simulation of physical or man-made processes.- A training component, where workshops would be given at various locations in the EU to familiarize spatial planning practitioners and students with the potential of recent technological developments.- An applications component, where various classes of spatial analysis problems are compared, and general suggestions and toolkits provided to assist in particular domains - e.g. forestry, oceanography, landscape management, and urban planning.
Оригинален текст от CORDIS (на английски).
Участници
- UNIVERSITY OF GLAMORGAN · PONTYPRIDD WALESКоординаторНиво държаваОбединеното кралство
Връзки
Данни: CORDIS, © Европейски съюз
