The Ask-It Approach
Dr. Angelos Bekiaris
DR. ANGELOS BEKIARIS: Thank you, John. Actually when I prepared the title of this workshop it was a bit more technical, a presentation on how we do the localisation of the service. Nevertheless, I will explain the concept and try to be a bit more generic as well. I have the benefit if you have difficult technical questions for me.
ASK-IT is an EC project. There is quite a big investment, it is 16 million euros investment; 9 million from the Commission, the rest from industrial partners. It's a four-year project. We are just in the first year.
You can skip the jargon. Very simply, we would like to use the devices like PDA's and mobile phones - so any current mobile phone, eventually with GPS, state-of-the-art and existing models as well - to support any user to go from a place of origin to the place of destination, and while he or she is doing that, to do the things we are all doing when travelling.
If I am in the airport, I can be connected to my business. A blind user who needs to have special software, a screen reader, might not be able to do it unless you provide a means by which he can enable the special software to work automatically while he is there, and go back, without having to install it.
So, not only do you have to be guided seamlessly from one place to the other, intelligently, personalised, but also you have to be able to do payments and other things on the move like everybody else. We have special groups in our mobility impaired group. Basically the special groups are not special at all. I am all of these.
I consider myself mobility impaired when I am carrying 50 kilos of luggage. When there are stairs, I do need to find the elevator because I have accessibility problems. It is important. When people are waiting for me, I begin to forget names and people have to keep reminding me. I speak four languages but, if I am in Osaka, I am illiterate. I couldn't find myself anywhere. You understand that I have the needs of specific types of people. Who is this? You and me.
This is an advantage of the services we have to offer. If I want to go from one place to the other, say from my home here, I need, first of all, to have in the model transportation information, meaning should I go to the airport by car or bus, or maybe I have to change bus. I haven't been to the airport, so I have to know which is my gate, if possible. Then I take a plane and come to London, and I have to find a way to go from the airport in London to my hotel.
All this, of course, has to be accessible for me. I might have a specific disability - for example, I need to book an accessible taxi - or accessible and personalised might mean I don't want to pay for a taxi, or I don't have the money, or I don't want a taxi, so other solutions should be given.
The point of interest, the venue. I need to know about the venue. I need to know about the accessibility of the venue. We have seen some very nice examples, for example, before we even reach London, like photos and videos.
Personal support services, where I can find assistance about public services and accessible websites, such as libraries, for example. If I am coming from here over to Greece, it would be very nice for me to see information about the Parthenon. I need accessible information, so this information should again be available. Social and community building. For example, a blind user or a deaf user might be interested to go to the local community to get information about particular events this community is doing, and so on. I just give you a very brief idea of the services.
Before I go to this specific model, I would like also to say something about personalisation. Personalisation in ASK-IT has three layers.
The first layer is static personalisation. Static means I put in my own disabilities, so the device should know if I have a wheelchair or even whether I have an electric or manual wheelchair.
Second, special medication, I may take things that the device cannot know if I don't put them in, but these must be limited because no user has the time or interest to put in big files.
Then we have dynamic personalisation. If you ask for information on a PC, you scroll and you get what you want. If you do that with your mobile or PDA, you can never scroll fast enough. The device can give you at best ten choices, so the device should give you information based on your location, your surroundings, but not too far. That's the first thing.
The second thing, is most interesting, according to your own preference. What we are doing is we are monitoring the preference of the user, in terms of how many transportation means are acceptable. So, when we tell you to go to a place, you might have to change four buses, are willing to do that, and so on. The system is learning about the user and, most importantly, you have the possibility to see this file and delete it or modify it, because no system is perfect, so the content is personalised statically and dynamically.
At the same time, the user interface is personalised. So, in the case you are blind, there is no reason to give you visual information on your PDA. We know that, so we put everything into tones, for example, or into speech. In case you have a visual problem, we can magnify the writing or the script. This means we can also use less lines, which means the content has to fit in those lines. If you are illiterate, you need to get icons standardised. That's what we are doing.
So, there is a very strong personalisation, it's a multi-intelligent platform, and one of the parts I am going to highlight today - you understand there are more than 50 packages and 5 interlinked sub-projects, so it's very big - is to look at the localisation and navigation and the route guidance parts.
First of all, you don't like to have a lot of navigation systems. You know today that there exist pedestrian navigation systems, and there exist in-vehicle navigation systems. In this case, you don't want that the system takes you in your car, and then you go to the end of your orientation and you go outside, and then you enter it again. We integrate this. Also, we combine outside navigation with indoor navigation, so it is in a way seamless.
Me as a user, I just want to tell the system where to go. If you want more you find it yourself. There are certain things I don't want to know. This gives you seamless navigation. This means a lot of things. How can we do it? With a level of accuracy up to 1-2 metres. Why? Think of navigation of the blind. To turn left, one metre is really near. You cannot say turn left and then there is no left. So, for many of our users, we have to be quite accurate.
We have to use standard interfaces. As I said before, we support integration of offboard and onboard navigation, integration with a car navigation system, the routing, and interfaces at different points around while you navigate because maybe some of them are relevant to you.
I am going straight to more interesting things. This is, again, I think, too technical, how we apply the system, and all the standards we are considering. It will be on the website, so there is no reason for me to go into very big detail on that.
For example, here is an interesting thing about indoor routing. Today the best possibility of doing the routing is with wireless LAN and WiFi, and we are using this. There are new technologies. The MOTE network is one of them. We are going to use more and more, even outdoors, the so-called Smartdust sensors. They are very cheap, very local, small communication sensors.
Actually the first application in transport they've integrated in some kind of paint. You paint the sensors inside on the street and the car goes over this paint. You get information like what is the road friction factor if there is rain; it will let you know. It is only 50% more costly than a normal paint, which is not cheap as a paint but it's extremely cheap as a sensor.
In this case, you understand that these technologies you can apply today, because in the airport you have already WiFi wireless LAN, so we can connect very easily. In other cases we need to mature this technology and to equip more and more, for example, train stations and other places. The idea is that you cover an indoor area quite well with this without really any special account of the user.
First of all, you need to import the maps. You have all the details there, and we are doing that. Most importantly we need the accessible routes on the map, because we cannot give up-to-date information unless we have the core information on accessible routing. That's what we are doing in some sites.
Here you can see the difference - they did some maps of today and they did some maps of tomorrow - not in ten years, but in three to five years.
We are working with companies who are providing the maps more and more. Here you can only see the street and you can only have a number. On the other, we can find out with much better accuracy, to one metre, which is the lane you are in, where is the pavement, even information about the pavement accessibility. This is what is actually being developed now for many European sites, and it will be more.
The ITS London event is happening in October 2006. We have there a demo. Actually we will apply our technology, our first prototypes, to take you from the bus stop of the venue to the conference centre, for different people with different disabilities, with more or less standard PDAs and mobile phones. This will be in London. We are building it now and October is quite near.
We also trying outside to put some more interfaces to satellite systems. As you heard, Galileo is coming in a number of years but, for the time being, we are upgrading the signal with EGNOS. This satellite system is an intermediate solution but it runs quite well.
MOTES, RFID tags and Smartdust are some of the technologies we were discussing before. As you know, the sensors are extremely small. They are extremely cheap as well, so that might be another technology which comes more and more.
I should also say here we are applying the logical navigation. What is logical navigation? You go into the bus. In the bus you lose your GPS signal. We cannot put sensors on buses. How do you go about it? If you know that the user would be at a certain point at the bus stop because his route planning instructed him to take this bus, and the user disappeared, you are bound to know that he didn't disappear. He is disappearing in the bus and coming back after a few minutes in the bus.
Clearly, if you are travelling at more than 70 kilometres per hour, you are not walking or running. But even that might not be the case in a city. In a city, you might be going 5 kilometres an hour and still on the bus, so we have clever algorithms by which we can do more than the sensors tell us. At the end of the day, it's always better to ask the users, "Are you there or not?" than leave them alone and lost, but this is certainly the last way.
So, we still have problems. Urban canyons, areas with no GPS signal are a problem. Also, some of the existing indoor solutions are the existing ones, like wireless LAN and WiFi, which are still expensive.
By the way, I am impressed by what is going on in the UK. Gateshead as well as Sunderland have WiFi on the lamp posts. We can offer some of that right now, but not the whole, of course. This technology will mature in the next few years. ASK-IT is not looking 25 years ahead. It is also not looking at the next year. It's looking at five years from today. Five years from today, it will be working in full, cheap, everywhere.
I would like to add here that some of the key partners of ASK-IT are Siemens, Nokia, Alcotel, Lucent, Microsoft, Vodafone, working on their standard devices. ASK-IT will be available at no cost at all for the end user because the manufacturers' interest is to make you use these web services, and the communication and services. It is not to sell you another software device. So, we speak about the mass market coming to you for free in the next few years; in the next probably four, five or six years at most.
I am not going to go too much on the technologies. There is the Galileo solution which will be ready for 2010. There is the EGNOS, which we are working on right now, and it is running. This will all be on the FTP. This is something that some of you might study, especially the most academic and technological partners, and we would like your feedback whether our solutions seem ok to you or not.
Up to now, we are going to install and run the system as from next year in these cities: Madrid and more or less the whole of Spain, Genoa and Italy, Athens and Thessaloniki, Bucharest, Nuremberg, Helsinki and Newcastle upon Tyne. These are what we call core sites, meaning for nearly all the user groups we aim to give nearly all the services in each core site. Then, there is the site in Den Haag that is looking at blind users, and some of the services.
Any provider who would like to connect to the services in any country through ASK-IT simply can use a relevant tool from our website to do so. It will take him no more than a couple of hours and the service will be connected. Basically, we try to make the service standard for users. We have connected already 80 service providers in Europe. By the end of the project, we want to have connected 300.
And here we see things that are happening in Sweden, in the UK, not only Newcastle, in the US, and we are talking right now with Brisbane in Australia and Beijing in China. We want to be connected to all; it is so simple. The devices and content are there; we can serve them. To connect our services for public use would be quite feasible - or we hope so - and certainly we would like to have you with us.
For this there will be a big conference on ASK-IT and a big exhibition. This is on 26 and 27 October 2006 in Nice in France, and our website is still open until the end of May 2006 for applications. We would be very happy to get papers and presentations from you.
Actually, for the academics here, and for those that have contacts, I can say that we have also two prizes. One is for the best students awards. For that paper we give not only expenses for the person to come to the conference and stay, but also 1,000 euros cash. And the same for the best paper from a developing country.
So, you can go and see our website. There are interesting figures. We have a lot of technologies. Also, I welcome any of your collaboration. Thank you very much.
APPLAUSE
John Gill: Thank you very much. To repeat, the website is www.ASK-IT.org. Any quick questions. I will allow one quick question.
NEW SPEAKER: This is not really just ASK-IT. It's relevant to every speaker that we have had today. It seems to me that so much information is needed if you are going to include all these different service bodies, every museum, every restaurant, every public toilet, every street. How do you propose to get the information?
DR BEKIARIS: Thank you. First of all, in several places, all this information exists. I will give you as an example Helsinki or Nuremberg or even Genoa, where accessibility information on every restaurant or any hotel exists.
However, it's not connected to any system, so the only thing you can get today is a booklet. Booklets are big and have a cost to be printed - you are lucky if you can find one - and in the other cities you have to find the new one. In these cases, it's more actually to make it accessible across the people than to make the information. On the other hand, there are other cases where we have no information at all. In that case, we build and collect at least the nucleus of the information. For me today, if I go to Thessaloniki and I am a blind user, there is no information whatsoever about accessible hotels.
Eventually, of course, we try to establish on each site different market organisations to make it viable at the end of the project so that it can continue. In each of the sites the market is totally different. In Madrid, the information is for free, because it is supported/funded by the Municipality. In Nuremberg, it's done by the Industrial Chamber of Commerce, because they say, "We have the money and we will do it, and we are going to sell this."
In other countries, they are maintaining this, connecting it and selling it with private companies. There are many different schemes, but we can only integrate what exists or the things we can do. We are not going to solve the problem of information accessibility everywhere, but those islands that exist - and they are becoming more and more - if we connect them, we have much, much added value. That's what we can do up to now.
NEW SPEAKER: Well, all I can say is I wish you the best of luck when you do London!
DR BEKIARIS: It's not a coincidence we have not selected it as a site!
JOHN GILL: Thank you very much. I think at that point we should move on to our next speaker.
| Contents page |
Last updated: 20.11.2009 © Copyright reserved Website design: Digital Accessibility Team