Science Spin May 2009

Moving down the road

By Marie-Catherine Mousseau

Whether we like it or not, cars have become a central part of our lives. Transportation systems and the routes they use have greatly influenced both how and where people live. In the face of increasingly populated urban centres, efficient road transport means increased mobility, higher speed and more freedom.Transportation is also vital to the economy - providing access to natural resources, promoting trade, in short allowing a nation to accumulate wealth and power.

Dark side

But there is a dark side. Transportation in general, and road transport in particular, increasingly has to face major challenges - environmental, energetic and safety-related. Road transport must respond to climate change and pollution concerns, rein in energy consumption and decrease dependance on fossil fuels, alleviate road congestion, and, last but not least, cut traffic fatalities (which cause 40,000 deaths a year in Europe).

"Climate change, increased transport volume in urban centres, transport safety, as well as a sharp increase in need for transport of passengers and goods, demand rapid and efficient solution," said Mr Janez Potočnik, European Commissioner for Science and Research at the opening of the 2nd European research conference on road transport. This statement illustrates clearly the slogan of the conference: "Greener, Safer and Smarter Road Transport for Europe."

That's what we urgently need; but how do we get there? To address this question, the Transport Research Arena (TRA) 2008, which took place last April in Ljubljana, saw an impressive range of experts - from automobile industries, to development planners, legislators, and of course researchers. New engines, alternative fuels, hydrogen technologies, smart cars, and emissions reduction were all part of the agenda.

Awards

Another part of the TRA agenda was to stimulate interest in transport research by spotlighting its most exciting promises. This was the remit of the YEAR (Young European Arena of Research) 2008 Awards. Funded by the European Commission through FP7 and coordinated by UCD, YEAR 2008 was a student competition recognising outstanding and creative transport research being carried out by students from all over Europe.

The scope of the event was significant. A total of 28 different countries participated in six categories of transport research. The Gold medal winners, announced at the TRA opening ceremony in Ljubljana, came from France, Sweden, the UK, and Ireland.

Ireland in fact did particularly well, as it was awarded the Gold medal in two categories: "Infrastructure Design & Production" and "Future Visions" - the category for creative ideas where applicants were invited to let their imagination fly and present their vision of where road transport will be in 20, 50 or 100 year's time... And Dublin city turned out to be a productive source of inspiration for the young candidates from Ireland.

Tommy, Francesco, Dónal and James' examples are showcasing road transport research in Ireland, but there is plenty more going on all across the EU. As Professor Eugene O'Brien of UCD Urban Institute Ireland, organiser of the competition, put it: "There is a wealth of Transport research being carried out in Europe's universities - by students from all over the world. Our PhD students spend many years developing new processes and techniques to make our roads safer, greener and smarter. It is a goal of the YEAR 2008 competition to ensure that they do not work in dark rooms by themselves."

He added: "This work is important, not just for the work itself, but also for the human resource that it produces. This new generation of transport experts will be the future leaders in the industry."

The good news is that there is plenty of room for everybody. Indeed, the transport industry is the largest industry in the world. From geotechnics (e.g. slope stability) to structures (e.g. bridges), pavements, transport planning and economics, automotive engineering, safety, energy and the environment, the scope is huge and the areas of research numerous to say the least.

But let's not allow the enormity of the task discourage our pool of young researchers. If they keep up the good work, then who knows better more efficient transport in Ireland and elsewhere may be just down the road.

Future visualization

We are quite familiar with the idea of visualising a landscape. "Visualising" a soundscape however sounds a bit more unusual. This may be one of the reasons why Francesco Pilla got a Gold-Medal in the category "Future Visions".

Francesco is an environmental engineer who got his degree in Bologna, Italy. In 2007, he started his PhD research in Trinity College Dublin with SFI and the Metropolis (Supercrowds for Multisensory Urban Simulations) group.

Like Tommy's, his project deals with urban and road planning. But this time it has less to do with optimising the traffic itself, and more with minimising its adverse consequences to ourselves and the environment. Air pollution would be an obvious traffic annoyance to think about; but there is another one maybe not so much publicised albeit very prevalent and annoying - that is noise pollution.

Noise pollution in cities has always been a concern, long before air pollution and even before the automobile was invented. As Francesco put it, "Julius Caesar is believed to have been the first to enact a noise ordinance, banning creaky chariots from barrelling around Rome during night time." Noisy chariots are no longer an issue, but as cars became prominent obviously things haven't improved. Some two thousands years after Caesar's noise ordinance, the EU has issued an Environmental Noise Directive (END) to help control and mitigate the noise problem. "The END forces competent authorities to prepare noise maps of certain areas," Francesco explained. "As a result, noise maps are now available for all the main cities in Ireland."

Another central premise of the END is public involvement. This requires that all information must be presented in a "clear, comprehensible and accessible manner," Francesco continued. Noise maps, however, hardly meet this requirement. "A noise map is a very technical paper with noise information on it," Francesco explained. This is basically a 2D representation of a city or part of it where decibel (db) levels of different areas are noted. "But for many people 50db doesn't mean anything," he said. "My main concern is to explain to people what 50db means."

So Francesco came up with an innovative and futuristic means of displaying noise - one which is more direct, engaging, realistic and intuitive, and also interactive. "What I'm doing is presenting realistic sound in a 3D virtual environment," he said. "The three dimensional representation of results is an ideal presentation tool for those who are unaccustomed to two-dimensional drawings. It allows a sensory experience."

Francesco's idea is then to show the generated noise map to people, to let them 'experience the noise.' "People would be able to walk through an environment - similar to a video game - and hear the sound," Francesco said (they work in coordination with the Computer Sciences department which provides the 3D environment, and even with the Psychological Department to predict crowd behaviour).

The approach is two fold: it is based on noise mapping and actual sound recordings. "The map is made by one of my colleagues and I'm recreating the sound." To produce the noise maps, they have their own software, where they can enter all relevant parameters - car frequency, data on the ground surface, building height. The programme processes the data and simulates the noise. At the moment Francesco has already simulated the 'background noise' in a virtual representation of an urban scenario surrounding TCD. The next stage is to add what he calls singular sources, such as a car passing by, people walking etc; so Francesco goes out recording the sound.

"The methodology has the advantage of being a truly interactive realistic environment and therefore offers a level of environmental realism that current noise mapping studies fail to deliver." he added.

"The model we are creating is web-based, because it's important to get through to the public and get them involved in the planning process - in line with the EU directive." Indeed, such a model would help politicians and the general public figure out the impact of the changes planned for a particular area. "If they want to build a noise barrier, you can include it in the model and hear the benefit that you would get from it," Francesco explained. It may also be used to highlight to people the impact of different traffic management scenarios on the sound environment. It may even help them choose their house by experiencing directly its surrounding noise without having to leave their computer.

The benefits of Francesco's technique could even extend beyond noise. According to him, it could be adopted in the future to include other environmental information such as air pollution. "Such a tool may fundamentally change the way in which planners communicate ideas and developments to the public - by modelling the built environment and being able to interact within it over the web," he concluded.

Since then Francesco started up a company in Dublin on noise consulting. "Although Infrasonic is a consultancy we are involved in research based projects. These aim to not only provide solutions to acoustic and vibration problems but also to improve the knowledge base of the field."

For more informtion click here

Aerial LIDAR

PhD student Tommy Hinks was awarded the YEAR 2008 Gold medal of the category 'Infrastructure Design and Production' for his research on Aerial LIDAR conducted in UCD.

LIDAR stands for Light Detection and Ranging. In many ways it is similar to RADAR (Radio Detection and Ranging). Like RADAR it is a 'remote sensing technology', but instead of emitting radiowaves, it emits light (shorter wavelength) in the form of many short laser pulses aimed in many different directions that bounce off the surroundings. Timings are converted into distance, thereby creating a 3D view of the surrounding. "Over the past few decades laser scanning has evolved into a formidable tool for rapidly and accurately recording 3D point data," Tommy commented.

Tommy applied this technology to "capture" Dublin - that is to create a virtual city model of the Irish capital. For this he had to go airborne. "It takes one day - 15 hours - to do one building from the ground, going all around the building with a scanner and a tripod," he said. "But if you do it from a helicopter, in one day you can do an entire city - that is thousands of buildings."

Tommy and his team didn't invent the technique of aerial LIDAR, but they improved it greatly. He explained: "Directly beneath the scanner it's very difficult to pick up a wall, because the angle is very oblique - basically, data beneath the scanner is dead. What we have done is design an overlap scheme which allowed us to transfer the data from good areas to bad areas. As a result sampling becomes more homogeneous." He added: "This means extra flight time but what we gain in quality is worth it." In other words, they optimised the path of the helicopter and the processing of the data. Thanks to Tommy and his colleagues we now have not just the roofs of Dublin, but also the windows, the doors, the pillars.

Tommy acknowledged that "the maths involved were quite nasty." He has a Masters in Computer Graphics (from the University of Linköping, Sweden - he's Swedish) which helped. His research also involved two departments at UCD, computer science and civil engineering. But everything at the moment remains confidential. "We had to hire a company to fly the helicopter and get the data, but they were not told what they were doing, they just knew that they were flying in a particular pattern," Tommy said. That's because Tommy and his colleagues intend to patent their technique as they have more ideas about how to process the data. "We are getting the patent; it's very close." he indicated.

But before that, they are expecting results on another aspect of their research. Beyond optimising flight paths, they are designing a new method of extracting the relevant information from all the data. "The amount of information recorded during a city flyover is staggering; reaching billions of points for a single city." Tommy pointed out. So what they are trying to do is to transform these vast amounts of point data into more visually pleasing and accurate models. "Because current techniques do not expect fully 3D point data, they fail to exploit this aspect." he said. "Points of different buildings tend to mix together - you don't know how the points are connected and which point belongs to which building," he continued. "In order to get an accurate visualisation we need the surfaces."

Promising progress is expected with 'geometric reasoning' which has the potential of identifying surfaces. "Using a completely new approach to identify vertical surfaces, we can identify building outlines in a more robust way than with previous methods," Tommy explained.

"So the next step of our research is automatically generate models like the ones in Google Earth," he continued. The difference is that with Google Earth the images are made by hand by individual users. The data there is not verifiable, and not necessarily homogeneous with regards to level-of-detail. "Imagine, with our technique in a few hours you get thousands of buildings, top quality, centimetre-accurate representation. It's automatic and much more efficient - an entire city in one day!", Tommy concluded. He hopes that in a few years time detailed digital city models will be available for everyone, from urban planners to game designers.

Body-in-white design

TOMMY and Francesco were actually not the only winners from Ireland. Dónal Gildea won an iPod nano for coming second in another category: 'Automotive Design and Production'. Indeed, optimising transport solutions is not just about optimal transport routes and planning. At the heart of the matter, there is the vehicle, and at the very heart of the vehicle there is the 'body-in-white'. "The term 'body-in-white' (BIW) is used within the automotive industry to describe the structural body of a vehicle," Dónal explained. It is the core, the 'bodyshell', where all the various components (engine, suspension, interior trim etc.) are attached. And it is quite sophisticated in itself. "To form the bodyshell of a medium-sized vehicle, the construction typically involves the need to join approximately 300 stamped sheet metal components together," Dónal said.

"This bodyshell represents the heaviest singular component of a vehicle and has the largest influence on many of its key attributes - including handling and crash performance," he continued. In other words the BIW is a critical component of a vehicle, a determining factor for its safety, but also for its energy consumption. This is simple physics. The lighter you can make it, the less energy it will require to move it. And the stronger, the safer.

So to combine safety and energy efficiently, you have to make it both light and strong which may appear as a challenge as to the materials used. "Environmental legislation and safety performance has provided the motivation for many recent developments within the field of BIW construction," Dónal explains. "In order to meet these objectives, car manufacturers are incorporating an increasing amount of high strength steel (HSS) into BIW structures." Use of HSS means that thinner, lighter components can be substituted for those thicker, heavier conventional mild steel ones.

The results is that more and more types of materials are used- mild steels, high strength steels, as well as extra high and ultra high strength steels - all included in a single BIW structure. This also means that a wide range of joining technologies is now required to assemble these numerous material types. "A typical car body will now contain any combination of spot welded, MIG welded, arc-welded and more recently, laser welded joints," Dónal explained. "Mechanical joining methods such as clenching, riveting, nuts and bolts are also utilised in numerous applications," he continued.

In fact, there are no less than 20 different joining technologies which on average are now required for the production of one BIW structure. This evolution has greatly complicated the task of the automotive designer. "Selecting the optimal joining technology for a particular component is at present a very time-intensive task which relies greatly on the personal experience of the designer," Dónal explained.

So Dónal thought about a systematic way of picking up the appropriate joint technology - a design methodology, as he called it. "When you design a car, there are many parameters to consider, and we cannot consider them all," he said. "So we tried to establish the minimum parameters that are needed to identify the suitable technology - such as joint length/type/accessibility, material combination/thickness, or degree of permanence. These are the key parameters." He continued: "That is the process side of it. Then we had to factor in economic data. What I've done is to establish the cost of making a standard join for each joining process and each material combination."

The result is a user friendly database on Microsoft Access which guides the designer all the way through the selection of the best technology, "thereby reducing design effort and development time," Dónal concluded.

Dublin bus

Dublin city was also inspirational to James Decraene. Even though he did not win any prize, James is still very happy he has been preselected for a project which happens to be more closely related to transport as well as a very much longed for map of Dublin bus routes.

James Decraene was born in France to a Chinese mother. He lived in the UK, then in Canada, before starting his PhD on artificial life in DCU, Dublin. But the reason James entered the YEAR 2008 competition had nothing to do with artificial life. James set out to draw all of the routes of the Dublin bus system. "The idea came up after my girlfriend had a bad experience using Dublin bus. She was to go to an interview in a part of Dublin she did not know. She eventually managed to find the right bus but could not figure out where to get off; she ended up missing her stop and missing the interview."

Unlike Tommy and Francesco with their Dublin 3D visualisation projects, James did not use any fancy technique or complicated maths. His idea was similar to the "Wikipedia approach". He set up an interactive website where fellow-Dubliners were encouraged to participate and send him the routes they knew. In a few months half of Dublin bus routes were already online, integrated with Google Map. The system even provides for the inclusion of additional data such as bus timetables or station locations. "The Department of Transport told me they intended to draw up a similar plan of all Dublin bus routes," James said. "But it's been nearly two years now and they still have not developed anything similar. There is today no online/offline map of the public transport network."

Surely, they didn't take the collaborative approach. As in Tommy's project, Dublin served as a pilot study to show that the concept works. James demonstrated that using a collaborative approach to provide optimal transport planning is efficient and applicable to other cities. What's more, behind this concept there is the potential to quickly identify 'hot spots' where transport is deficient. "Conventional systems have a significant lag time in identifying these problems," James explained, "our system is highly reactive, so it can intervene faster before things go out of control."

He added: "I think the important idea here was to show how through a collaborative approach, critical data could be obtained - and not only bus routes. This approach could be used to gather other information, e.g. traffic, to do efficient transport planning and conduct further studies."

James may not have won the final prize, but his achievement is still worth noticing. He had no funding and did it in his spare time. With almost no marketing he's getting around 10,000 unique visitors per month.

For more on this project click here