Bridging the Worlds of CAD and GIS - Part 1 of a Series on CAD-GIS

By Peter van Oosterom

Information systems, which involve geometry, are used for many different purposes.One could classify 'Computer Aided Design' (CAD) as one such family of systems; this, in turn, is often related to another, namely: 'Computer Aided Manufacturing' (CAM).The products that are designed and/or manufactured consist of movable objects (tables, cars, airplanes, engines, coffee machines, electronic circuits) and fixed objects (plants, buildings, houses, railways, roads, bridges, tunnels, utility networks).CAD systems for fixed objects are applied in AEC fields (Architecture, Engineering and Construction).
Fixed objects are also well-known from another family of information systems, namely GIS (Geographic Information Systems).GIS is applied, amongst others, in urban planning, land use and cadastral data handling.CAD and GIS share one major characteristic - both deal with geometry - but they differ in many aspects (size, storage, analysis, semantics, attributes, etc.).So, why bridge the gap between the two? There is a growing tendency to integrate them in applications.This can be explained by the fact that CAD and GIS systems provide information on and deliver representations of the same real-world objects in each phase of the lifecycle.There are several areas of application that illustrate the need for an integrated approach:
  • Plan development: The design of large infrastructures (roads, railways, bridges, tunnels, etc.) needs both CAD and GIS information: CAD techniques are applied for the engineering and construction while GIS data is essential for the initial planning and layout.In the design phase the geographic description of the region is often transferred from a GIS to a CAD system. Once the design has been completed in CAD, it is re-imported into GIS.
  • Visualization: Plan presentation and data interaction often require different 'views' of the data: a 2D 'plan view' for the initial context analysis (GIS-like), a 2.5D 'model view' to create and evaluate the different design concepts (mixed CAD/GIS-like), and a 3D 'world view' to realistically visualize the subsequent design (CAD-like).
  • Data collection: In recent decades, data-collection techniques have progressed from manual measurement to remote sensing and photogrammetry (automatic interpretation).Some advanced photogrammetric techniques assume knowledge about objects, such as buildings, bridges and other landmarks, in a CAD-like format.Also, the difficulties of surveying sub-surface constructions (which cannot be surveyed from the outside), are fueling an interest in CAD models in 3D GIS modeling.
  • Location-based services and augmented reality also employ a combination of CAD and GIS techniques for positioning, deriving viewing directions, and supplying the user with relevant 'sight' information.It will take a lot of GIS and CAD integration before a sentence like "On your right hand you now see a 12-story building" is generated automatically by the computer or when a planned building (CAD model) is superimposed via a special glasses on the true real world image.

These applications require functionality from both worlds.So, why are these worlds so difficult to bridge? Essentially, because CAD and GIS traditionally have a different focus:

  • CAD represents the man-made world while GIS also captures the natural environment. The underlying mathematical description is therefore quite different.Whereas CAD represents single complex objects in 3D with a high degree of accuracy (including free-form surfaces, etc.), GIS aims to capture large numbers of objects in a common embedding.
  • The timescale is quite different.As CAD generally works on a 'project' basis, lifecycle maintenance is a fairly recent issue.GIS, on the other hand, is geared to a very long period of data collection and maintenance (almost an endless lifecycle).Whereas CAD often stores data in a file format, GIS more often maintains data consistently and permanently in large databases.
  • CAD systems generally assume a (2D or 3D) orthogonal world, while GIS systems deal with data sources based on many different coordinate systems, which are used to model the spherical (ellipsoid or geoide) world.

Although CAD and GIS information relate to the same real-world objects, the data is quite different and takes account of quite different aspects. Also, all these different pieces of information are created and maintained by totally different organizations with different tools, optimized for specific tasks.In the late 1980s and early 1990s, many papers were published on GIS versus CAD and on how they could be effectively combined.However, these papers tended to focus mainly on 'how to use CAD systems for certain GIS tasks', ranging from geographic data entry to automated map production. About ten years ago, inspired by application domains, such as urban and landscape architecture and planning, attention turned to the integration of CAD and GIS functionality.But the solutions were ad hoc (capturing and transferring simple 3D models between the different systems).The fundamental problems behind the integration difficulties were not addressed.

A true solution for an integrated CAD/GIS framework should cover two lines: formal semantics and integrated data management.The development of formalized semantics is crucial to achieving the true integration of CAD and GIS.First, the semantics (of geometry and other information) within a domain need to be formalized, i.e.domain ontology has to be developed. Next, these domain ontologies have to be matched against each other in order to have meaningful exchange of information between the two worlds. This could be realized through an integrated (and refined) ontology covering the CAD and GIS concepts in one framework.

After solving the semantic differences, the next step is to create an integrated model that can serve multiple purposes (from both a CAD and GIS background).Different views may be defined on this representation. The integrated model is managed in a way that maintains consistency during updates or when model data is added to the data base management system (DBMS).So, the same model is used as the foundation for planning, design, construction, management, analysis, presentation, and so on.Shared data management does away with conversions and all the accompanying problems as everyone taps from the same source.DBMSs also offer other well-known advantages: multiple user support, transaction support, security, and so on.To summarize, 'islands of automation' will be abandoned, as company-wide (including CAD and GIS) information management becomes a reality.

Other installments published to date in this series:
Introduction: Large-scale 3D data integration - An Introduction to the Challenges for CAD and GIS Integration
Chapter 1: Bridging the Worlds of CAD and GIS
Chapter 2: 3D Data Acquisition and Object Reconstruction for AEC/CAD
In addition to the articles themselves, numerous readers have also posted comments on them (found at the bottom of each article).

Published Friday, June 18th, 2004

Written by Peter van Oosterom

If you liked this article subscribe to our bimonthly newsletter...stay informed on the latest geospatial technology

Sign up

© 2017 Directions Media. All Rights Reserved.