|
|
|
|
|
|
|
|
|
| CSIRO | SOLVE | Issue 10 | FEB 07 | |
|
| ||||||
|
ARTICLE
MATHEMATICAL MODELLING:
Planes Apart By Melissa Marino
With airspace increasingly crowded, mathematical modelling is identifying the traffic control 'gaps'.To understand risk in air traffic safety, think of having a keyboard with a faulty 'z' key, says Neale Fulton, CSIRO's research leader in Optimisation in Air Traffic Management. "If you typed up a script where you never used 'z' you'd say your keyboard was working perfectly," he says. "And then one day you get a new script and you have to type the word 'fuzzy' and it fails twice in a row." Whether your keyboard succeeds or fails will depend on its pattern of use and Dr Fulton says it is the same with airports. He says to analyse risk in airspace you must assume that everyone is typing a different script: “All those different scripts are different use patterns at each airport. You have to understand the pattern of use at an airport before you can ever talk about the probability of failure, and that’s what we showed.” Airspace is increasingly busy and complex, with a mounting variety of aircraft in the sky following a set of air rules, much like road rules, under a range of operating systems. Systems governing major airports for commercial flights operate in centralised ‘managed airspace’, but as more general and recreational aircraft as well as uninhabited aerial vehicles (UAVs) take to the skies – and not necessarily with a flight plan – air traffic management becomes more problematic. In a benchmarking case study, Dr Fulton and Dr Mark Westcott from CSIRO Mathematical and Information Sciences were called into Broome International Airport to apply their skills in precision navigation and sophisticated airspace modelling, to evaluate flight-use patterns and support quantitative risk assessment. With a growing international tourist hub that is increasingly mixing Boeing 737s and general aviation, Broome airport’s management wanted to know what their safety risks were and what systems were required to manage them now and into the future. Using unique analysis incorporating advanced mathematics, computer science, aerospace engineering and statistics, the team modelled one million different collision geometries to test whether pilots had enough time to communicate and manage radio frequencies to avoid a collision. “The results of that study was that there were 50,000 geometries where it would fail,” Dr Fulton says. The research convinced Broome airport management they needed a ground-based operator in place to ‘fill the gap’ when pilots had to change frequency to stay in communication with each other. It also showed that an operator could work adequately from a site at ground level rather than in a costly control tower, saving several million dollars in capital and ongoing operational costs. Professor Stephen Emery, co-director of specialist engineering consultancy Kubu, was charged with advising Broome airport on the project and says he called in Dr Fulton’s team because they were the only people in the world doing this type of research. “It’s groundbreaking stuff,” Professor Emery says. “It provides, for the first time, an engineered approach to designing airspace. “While a lot of good rules and regulations were introduced after World War II, this gives us the engineering ability to design a system as technology moves on and airspace gets more crowded, rather than rely on rules created for a different environment.” Dr Fulton says it is important that airspace research keeps up with advances in aircraft engineering. “What we’re saying is that airspace design is just as critical as aircraft design, and we need to migrate the engineering to more realistic models, because the complexity of what we have to deal with is enormous.” Dr Fulton says airspace is destined to become even more complex, particularly in regional areas where there are increasing commercial flights and more general and recreational aviation. Then there are the UAVs – an array of unmanned craft that could be as small as a bee or as large as a 747 – which could be used for such tasks as cargo transport, crop spraying or geological data gathering, or even by real estate agents for aerial photographs of properties. Once, 95 per cent of aircraft in Australia were aware of each other, but today that figure is much smaller, Dr Fulton says. About 3000 commercial aircraft operate with good communications, while many of the 9500 general aircraft may be without pilot-to-pilot communications. Dr Fulton says while it is ideal for pilots to communicate with each other to exchange position information, a combination of systems is optimal. Ground staff add a strategic benefit at busy times when pilots could lose touch with each other, while the historic ‘see and avoid’ technique is still vital, he says. “We’ve looked at the communication systems and the kinematics (the position, speed and acceleration aspects of the aircraft) and what we’re saying is you can’t model airspace on kinematics or communication alone – you’ve got to bring those two things together.” Dr Fulton says advances in transmitting GPS data between aircraft would improve aircraft proximity management in the future. Automated Dependent Surveillance Broadcast (ADS-B) technology – known as data-link – speeds communication by transmitting position information digitally and synthesising it in the cockpit. CSIRO developed mathematical models to support the introduction of ADS-B technology into commercial aviation in Australia when it was first trialled in 1995.
Professor Emery says Dr Fulton’s work has provided important tools that can be applied at any airport to model complex systems and simulate them mathematically, instead of trialling them in the real world. “It’s foundation research as we advance the way we calculate, model and deal with risk,” Professor Emery says.For further information contact: |
 IN THIS ISSUE
|
| Home | About Us | eSubscribe | Links | |
|
|
||
|
|
|
Back to main