Modern technology and surveying instruments are continually changing the way spatial professionals collect and present data. Technologies such as Aerial Laser Scanning, Satellite Navigation Systems, Terrestrial Laser Scanners and Reflectorless Total Stations have all increased the productivity of the surveyor and the amount of data he/she can efficiently collect. Additionally, modern computing power allows this data to be displayed with unprecedented realism, allowing virtual 3D models to be constructed together with fly-throughs, scene views and user-controlled walk-throughs.

This thesis project applied this modern technology in order to survey and digitally represent Sydney’s Oran Park Raceway. As the site is to be closed and re-developed, the virtual preservation of the site allows viewers to explore the site in its pre-demolition form, and also allows users to make use of the track in Simulation Racing Software.

The intention of this thesis project was to create an accurate data set of the spatial information collected as well as a visualisation of this data in the form of a 3D virtual model which can be used to inquire pre-development spatial information, to create fly-throughs and virtual scenes, and to be used for driving simulators. Due to the large amount of work that was required, Brendan Elliott and Chris Larmour shared the workload throughout the project.

The authors would like to thank the following people for their assistance during the course of this project:

• Thomas Yan (UNSW) – For his assistance with the operation of the RTK-GPS equipment and Leica Geo-office.

• Brendon Pywell (Bob's Track Builder) – For helping us optimise the terrain and the track for the walkthrough application and rFactor export.

• Jeff Boulous (Manager, Oran Park) – For allowing access onto Oran Park Raceway and for his time and patience in organising the logistics of the field work. Also for putting up with the authors drilling holes into his concrete.

• Mark Perich (Owner, Oran Park) – For his support with the project.

• Mark Hitchenson (Camden Council) – For enabling the authors to get in contact with Mark Perich.

• Vicki Brown (Camden Council) – Providing advice in regards to obtaining data.

• Michael Andon (Growth Centres Commission) – For providing the Aerial photo.

• David Turton (AAM Hatch) – For providing the ALS data set, information as to how this data set was collected and processed, and for providing permission for use of the ALS data set in the event of a public release of the track for use in commercial simulation software.

• Glenn Morrison (AAM Hatch) – For his assistance with trying to provide a Laser scanner, and for his valuable advice.

• Michael Nicholson (BHP Billiton) – For his assistance in trying to obtain a Laser scanner for use in the project.

• Dr Bruce Harvey (UNSW) – For providing advice in regards to processing and Fixit3 questions.

Field Work

The fieldwork that was carried out in order to create this model was done in order to verify and supplement the ALS (LiDAR) data that was kindly supplied by AAM Hatch. This involved a control survey in order to establish primary marks around the site and tie into the existing NSW cadastral system and thus connect the survey to the Map Grid of Australia. Once this was complete the survey of the track could be carried out.

To do this, Real Time Kinematic (RTK) GPS was used, which allowed accuracies better than 30mm. In order to collect the majority of the points on the track it was determined that mounting the Receiver Antenna on a vehicle and auto logging the data would provide the quickest and most reliable method of collecting points. However, there were still a number of things that needed to be surveyed where car access was not possible or would have been inaccurate and these points were collected on-foot.

After the track and terrain data was collected and verified, the building survey took place. This was achieved using a Reflectorless Total Station (Leica TCRP1203), which enabled the positions of the key parts of each building to be determined – heights, lengths, widths and important features were all surveyed allowing the accurate development of the 3D models later in the project.

Modelling Processes

Various modelling techniques were combined to create the end result as the different physical characteristics of each subject item required different methods and techniques to model depending on the method and the equipment used to gather the data. The major techniques utilised were: flat roof, triangulated irregular network and constructive solid geometry.

The flat roof technique is the simplest method for modelling basic buildings. It involves locating the footprint of the building and simply extruding the footprint polygon to the measured height of the building. This is a very basic way of achieving 3 dimensions but it does not account for any non verticality of the walls of the subject buildings.

Another technique which was incorporated is a triangulated irregular network (TIN). This method was used to model the track and the ground surface surrounding the track. A TIN is capable of representing terrain in 3D using nodes and lines with 3 dimensional coordinates. In the case of thesis, the TIN was formed primarily from a combination of Light-Detection and Ranging (LiDAR) data and Real Time Kinematic (RTK) GPS observations.

To model the control tower, the bridge, the pit buildings, both grandstands and the various objects and buildings around the track, the constructive solid geometry approach was utilised. Using this method a complex structure is created using simple building blocks such as planes, spheres, rectangles and cylinders. Boolean operations are used to join the objects and to cut sections out of each model. Models created using this method require much more time and effort to create, however a better end result is achieved. Constructive solid geometry is a balance between over simplifying a model and forfeiting all of its accuracy and creating a model which today’s computer processors will cope with.

Final Model Applications

The final model can be used for a number of different applications due to its flexibility and large amount of data. Several uses for the model were explored as part of this thesis project, including the creation of a high accuracy longitudinal section of the centreline of the grand prix circuit, a user-controlled walkthrough, and the export to a commercial racing simulator (rFactor). Additionally, such a model could be used for shadow calculations, view analysis, engineering calculations (road and stormwater design) and possible re-creation of the track, buildings or parts of them. As well as these uses, the model forms a valuable source of historical information due to the sites pending housing re-development.




Chris Larmour

Brendan Elliott

© C. Larmour & B. Elliott 2008