 I'm really honored to be here and to obviously conclude this lecture series for the year here at this esteemed institute. And to talk about my favorite topic, the future of the automobile. And as I put the subtitle in there, there are many different facets. And I'm going to talk a little bit about all of those. And I'm definitely looking forward then to more in-depth discussion maybe afterwards. I want to talk about autonomous connected, electric and shared. I think the first three attributes are the ones that I'm going to highlight most and maybe get to talk a little bit about car sharing as well. And whenever you see this little Stanford logo, little Stanford tree there on top, then this should remind me that this is where I want to talk a little bit more about the Stanford specific approach in this field. So I want to talk about general things, but certainly I want to tell you what Stanford is actually doing about those challenges that we have regarding the automobile. I assume that many of us like the automobile, quite a few might love the automobile, some might hate the automobile. And there are a lot of reasons for each of this. The automobile can be potentially dangerous, can be dirty, and can also be very expensive. So these are just the numbers that we actually maintain in the United States when we actually want to talk about our motivation, why we are working on autonomous vehicles, connected vehicles, electric vehicles, and so on. So in the United States almost 33,000 people get killed every year in traffic accidents. It's an astonishing number, it's a very sad number, and of course you can do the math, how many airplanes crashing that would mean on a particular day. On a global level, I want to give you that number as well, and 1.2 million people die in traffic accidents every year. And 95% of this is due at least in part to human error. So we are not very good at controlling the vehicle, we are not very responsible. And the question might be how can technology maybe help with that. The other thing is congestion and pollution. Pollution I do understand, and I'm certain that this is something being discussed here as well, how many gallons of fuel actually these automobiles and trucks as well burn every year. But with congestion I also wanted to say there's one number in the United States that about one entire work week is being lost per commuter every year because the traffic is not as efficient as it could be. So the entire commute is actually longer than this, it's about 100 hours average per year per commuter in the United States. But almost about one entire work week people lose because traffic is congested. So what can we do? What solutions do we have on hand? And these are the ones I'd like to talk a little bit more about within the next half hour maybe. Driver assistance, we will be talking about autonomous vehicles in that regard. Connected mobility, what if we can connect vehicles the same way we can connect cell phones, computers and other devices. Alternative energy, certainly with the huge topic electric mobility in there, but also let's be realistic about all of this because many of these things we hear over and over again we say let's just do it, but the question is what does it mean? And I think this is where we need to take a closer look to do it right. In mobile society I suspect we might not have enough time to talk about this in greater detail but in order to make this picture complete, and as you can tell there are some pieces missing in there, mobile society I understand is how do we use, how do we consume mobility? With all of this I'm mainly talking about personal mobility, so the automobile, certainly public transportation is very important as well. And these things are maybe mobile society, how can we have an integrated and intermodal mobility approach connecting this with public transportation, but also ideas like car sharing, also what's actually happening with the demographic change. The younger generation has maybe a different identification with the automobile, different than my generation maybe had when we got our driver's license like 20, 25 years ago. And also this is thinking about the elderly who every or many societies obviously experience in change in their demographics and people are staying mobile, want to be mobile even longer and therefore might have different needs for mobility as well. And there you can already tell might need some more assistance, so there's certainly a lot of connections between all of these topics. But really what makes this, it's obviously a puzzle, what makes this puzzle complete are the different disciplines and this is where we actually had a very nice discussion already over lunch. Of course you can put technology in there. You can say I give you the technology and it'll all work. I'm a mechanical engineer myself, I like to work on these things, I like to find a great solution, but it might actually not really be the ultimate solution in terms of many other disciplines and this is why we need to look at economics and business obviously as well. There needs to be a business case behind this because otherwise people might not buy these things and also might need to address other topics in the field of economics. Sociology and psychology, in the end people must want to buy these things. As long as we don't mandate a certain technology, as long as we don't really declare everyone shall buy this technology then there must be a connection obviously between business economics and the whole sociology and psychology of the field so that people subscribe to a certain technology. But also and this is something we will discuss certainly with the driver assistance topic, do people trust a certain technology? Do they trust that the vehicle can do things better than themselves or is it something that they trust a certain business case? Will this battery actually have a good resell value when I sell my car in three to five years from now? The technology might be great but these questions might be not answered in a positive way. Politics and policies, we also had a great discussion about this over lunch and there are many different facets. I actually put this on purpose close to driver assistance because as we are going to be talking about autonomous vehicles, the question is what happens if something happens? If the car is smart, why is it then so smart that maybe still an accident happens as we obviously all experienced in March this year, there is no technology 100% safe and so accidents will happen whatever we put out there. But if we say the car is autonomous, I didn't do it, the car did it. What does that mean? And then finally of course resources and environment. In the end there is not always enough advocacy for the environment and talking for the challenges that we see there. This is kind of the picture how I see it and also how I actually instruct our students. The class that teaches the future of the automobile is actually located in mechanical engineering but I really educate students and encourage students to go beyond their standard or their traditional disciplines and look at many more topics. So from here we could go in many different directions. I propose that we maybe start out with driver assistance and connected vehicles and then talk a little bit about alternative energy and electric mobility and then maybe have a mutual discussion later on or to the ideas that you are bringing here to the panel today. So driver assistance, what does this mean? On this picture you actually see and this is why the tree should come on, no the tree does not come on, but on this picture you actually see the vehicles that we worked on at Stanford University and I assume that some of you are familiar with the history of autonomous driving at Stanford. On the far left you see Stanley which was the Stanford Entrant into the Grand Challenge in 2005 and Stanford together with Volkswagen, the industry partner was first, actually the first vehicle that won an autonomous race ever. Carnegie Mellon was second and to jump ahead to the picture in the middle two years later when we had the urban challenge, Carnegie Mellon was first and Stanford was second. So there was a lot of competition about these topics and I'm more than happy to talk later about how DARPA, so basically the research arm of the US military got interested in that. But also what we discussed very often, military might be the driver behind certain technology whether we appreciate this or not, but certainly there are many more connections maybe as I could fit in my little puzzle piece. And some of you maybe heard also about the car here on the right, which is nicknamed Shelly after Michel Mutant and the racing driver, female race driver obviously in the 1980s and drove up Pike's Peak in an astonishing speed. And so there's some research going on at Stanford University as well. Can you actually mimic a race car driver by means of artificial intelligence? The question might be, why do you want to do this? Why do you want to have an autonomous car? But in all of this you must understand it's certainly research and it's to show what technology is capable of. Again, technology is one puzzle piece and then further we need to refine those things. And then now that we have this technology on hand, how can we actually use a very competitive technology or a very competitive approach such as racing to make vehicles safer and actually have the computer take over from the driver where the computer is just better and has proven its capabilities in those races, for instance. So what do you mean by autonomous cars? And here is above this little line that you see towards the bottom. Basically how an autonomous car works, it gets a lot of input. It certainly uses GPS technology. But what's really important here is this part here. You have a lot of computer vision, whether this is a laser scanner or actual rear cameras or radar should be on this picture as well. That the vehicle tries to perceive as much of its environment as possible. And then it certainly also knows a lot about the vehicle motion. In the end, everything is being fed into the central processing unit and then miraculously you operate steering brakes and accelerate a paddle for European part car I might put in as a clutch as well. But basically the car can drive itself. What does this actually do? We talked about 95% of all traffic accidents, human errors at least a contributing factor. So the assumption is a computer is never drunk, it's never sleepy, it's never on the cell phone, doesn't have bad vision. Or maybe, might have bad vision, it's probably not drunk. But it might have bad vision and so you see already, who keeps these camera lenses clean? And that is something I also want to leave you with. Very often we hear here comparison to airplanes. Planes can land itself autonomously in bad weather and all of these things. Yes, because the pilot actually checked the entire plane before takeoff. But when did you check your tire pressure last time? You don't have to raise your hand or something like this. I just know that you did not do it this morning, not all of you. So there's already a challenge there. But definitely we can reduce the risk of human error. We can improve traffic flow. This picture, I believe it's from the United States. You find these pictures everywhere. Actually, we found this picture last night when we got on a taxi to our hotel downtown, the same everywhere. We are just not very good at balancing out maybe our travel schedule that we say I maybe travel half an hour later so that I don't hit the bigger traffic. Because you don't know if it's really going to be better. But you can imagine if there's some sort of intelligence that really balances the traffic flow and also the velocity of the vehicle. Because those of you who might have looked into flow theory, there's obviously a lot of science behind congestion. You actually do know that if everyone just went a little bit slower or just controlled his or her vehicle a little bit more accurately. And we might not have that much congestion. And improve efficiency of vehicle control in general. That basically means if you have a very nervous right foot, you might not end up with the specs that you get from the car dealer who says, well, I sell you this car with five liters per 100 kilometer and you end up with 6.5 or something like this. The car might not be wrong. Not saying that you might be wrong. But a computer might be able to do this actually better. So just a few examples of what we can gain here. And this is where we actually do a lot of work, where we improve this algorithms further. Very often I get the question, okay, the urban challenge was now four years ago, so what's next? So we're working a lot on pedestrian recognition, which is not just there's a pedestrian Volvo, for instance, has ready these systems in the vehicles. But it's a lot of difference if a pedestrian stands like this or like this. Because if here's the curbside, not many people actually walk backwards, they actually walk forward. And therefore you want to predict what is going to happen next. And this is where a lot of computer science actually can be applied. So the questions is, and this is a slide that I have for the other two fields as well. What are the key questions we are trying to answer? And this was also something that we discussed over lunch. Very often we are jumping actually ahead to come up with solutions. We don't always think enough about the question, especially the question we're trying to solve, the problem we're trying to solve. So what we need to find out here, is it actually technology that can be affordable to mainstream? Or is it just outrageously expensive? Will people feel comfortable with this? Do they trust this? And then an autonomous car is nice and well. But what I said earlier about the autonomous autopilot on a plane, they're actually too highly trained and very well paid or relatively well paid professionals monitoring the system. Is this what you wanted from an autonomous car? And then what does the mix look like? We can probably do something like deploy autonomous vehicles today, but there's still a lot of human controlled vehicles out on the road. Do they actually mix? And then do we need infrastructure for this? Obviously all these questions are connected. You only trust it if it really works and it might only work if you have a lot of infrastructure. But if you need a lot of infrastructure, it might cost a lot. So, and this is really where I say the vision is great. I'm a huge subscriber to that vision because it can solve all these challenges that we, or at least help with those challenges that I pointed out in the beginning. But the solution might be much, much more complicated. So one slide maybe about each of these questions. This is a study that students did in one of my classes. They looked at the cost actually that for those different vehicles need to be accounted for what we call advanced driver assistance systems. This is something like lane departure warning, collision breaking, parking aids, blind spot detection and so on. And you can tell that in some cases, if you really load your vehicle with all of these things, you add another $10,000 to it. No one has a price tag today for an autonomous kit that you can get on your car. But I guess we can assume it will not be less than $10,000, especially once we're talking about a few years out, inflation and so on. So, and what do you actually get for it? And that is really the question, can people trust it? I mentioned that earlier, the human will always be part of it because a human actually needs to say one way or another, yes, this vehicle is ready to go. I checked the sensor. I did not ignore this warning light. And I'm using that car as it's supposed to be, which might require actually reading the manual, which not many of us do. So there are a lot of things that we need to consider. And this is something we also wanted to highlight some differences between the United States and Europe. In the United States, many people expect this technology to be available very soon. Probably to some extent because Google is working on it, whatever Google does, especially if you start to hear about it, it's basically available within a few months. Now, the development cycle in the automotive industry and the development cycle in the IT industry is somewhat different. But still, it's expected to be available very soon. And it's supposed to be a relief from the commute or to something we discussed over lunch. Now, my observation in the European Union, I'm from Germany, as you know. So I have maybe a specific German perspective on that one. But it's really being seen more as an evolution of the existing. So more and more of these things that you saw on the earlier slide, a better adaptive cruise control, a better lane departure warning, a better vision system as an assistance system. And maybe eventually then the jump to a fully autonomous vehicle, but still more as an assistance than the car does everything, which maybe is a trust issue in itself. So it's very interesting for me to observe actually how data privacy is handled very, very differently in the United States and in Europe. And with the internet, obviously there is just the United States and Europe. But obviously everything is available everywhere. And it's always very interesting to follow actually what is now being banned from Twitter or what is now being adapted in the settings in Facebook. And that is certainly also something that needs to be considered with autonomous vehicles. Well, I didn't do it. The car did it. Well, did you do the fine print in all of this? This is where we actually do a lot of work. We do simulator tests where we have subjects actually driving a simulated vehicle, maybe in a standard setting as a conventional vehicle. And then in comparison as an autonomous car we question then the subjects, what they like about it. If they trust this car, they would buy this car. And we also do work, excuse me, in this field in the legal aspect. Because it is something in the end, if something happens, who's responsible for this. And this brings me to this picture, which I just have to put it in there every once in a while. It's a book that my parents bought in the early 1970s. And the book is called The Future. It's a German book, so it's called Die Zukunft. And it talks about the year 2000. There you have it. They thought in 30 years from back then they would have the autonomous car. Well, there isn't something like this out there, obviously. But the interesting notion of all of this will actually be, or will it be possible to turn drivers into passengers? And again, this is a huge legal question that we need to answer. And this is something at Stanford where the industry actually encouraged us to take a closer look into this topic. And this is what I do then as my main job. I take this closer look at this and then we actually create new topics in an interdisciplinary way. Because there was no one actually working on this at Stanford. But then we said, okay, it's not quite engineering. It's not quite at the law school. What about in between? And now we hired actually a researcher who's at the intersection of law and engineering answering those questions. The other thing why I always put on this picture, it all looks so nicely. So everyone is in the lane. There's nothing on the street or on the road that can be a problem. Real traffic looks different. There are many things that can happen. And can you put all these things in a computer system? Right? But we know that all these things happen. There might be a chair. There might be the chicken. There might be the other side of a horse. The question is, what does the computer see? And if you go back to, well, we would need to go back actually when we opened our eyes for the first time what we saw. We didn't know what was going on at all. But even when we got our driver's license to actually get our mental mind into something like what you see and how you should actually categorize something and then react and all of these things. So it says, it takes about seven years until the human brain gets really into driving mode. Sure, you can code all of this in a computer. But then you have it in the code. It does not necessarily evolve that much. We do quite some work on machine learning and things like this. But what happens the first time as a machine has not learned the situation? And then the driver's just a passenger and who's liable for this? So it's a huge complexity. And then the other question is actually, will it work without infrastructure? The word autonomous implies something. Some sort of independence. Don't need anything else. You just do it. But you might need something like vehicle-to-vehicle communication. Vehicle-to-infrastructure communication. Some of you I assume have heard of automated highways where you might have little magnets in the roadway or some other transponders that you know where the vehicle is in the lane and then you control the vehicle accordingly. Or the cloud, that you just get data from the cloud. Now the researchers who are working on autonomous cars, I can tell you they are not a huge believer in infrastructure. They basically say we've got a plow ahead, we have to do this. And honestly, I'm torn. Whenever I have these discussions with people from industry, with people from academia, also with people from the government such as Department of Transportation in the United States, I change my mind slightly every time. And because it's really, you can do so many things in this regard without an infrastructure. And just the other day, I had a great discussion with a professor from Princeton University who had the energy efficiency research at Princeton University. And he said, you know what, the traffic from Princeton to New York is just a nightmare. Living in Princeton is great. Working in New York is also very nice because you make big money. But in between, you earn about $1,000 an hour and then you're stuck in traffic for 36 hours a year? It's quite some money. So they basically say, you know what, we should just purchase one of these lanes, just close it and only allow autonomous vehicles on it. And that's maybe possible. So this is really where the whole deployment scenario and the transition from today's world and traffic into tomorrow's highly automated traffic is a very, very interesting question. I think the technology is there and whatever we need from this or might not need from this, we can do that. But how we transition there, big question. That's kind of the part I wanted to leave you with for now for autonomous cars. This is actually a very nice transition to connected vehicles because this is where we actually need all this infrastructure. And I will summarize the two fields after this part. Obviously for connected vehicles, which means vehicles are talking to one another, like the vehicle to vehicle, talking to an infrastructure, talking to the cloud, the internet in general. Obviously you do need a lot of infrastructure for this. I put in here different applications. So it can just be car data, like position, velocity, heading, vehicle identification number and all of these things that is being transmitted so that you can avoid accidents. I'm here at this position with this direction. We have a problem. You can do this with technology. A lot of challenges, but it can be done. You might get warnings what's happening out there on the road so that you know early enough what and when you might expect. Trip information, which is just your time of arrival, your route data and so on. And certainly we have a lot of entertainment things down here and yes, you should look at the internet and yes, you should look at these things together because the business case, especially for the vehicle to vehicle communication is incredibly challenging. It's just something people and the interesting thing with the automobile, we talk about people, we are talking about all of us in fact. People are not willing to pay for an infrastructure just because it makes it a little bit safer. There are studies basically that say this, I find this surprising, but probably if push comes to show that's the scenario. But they're willing to pay for infrastructure for media. Willing to pay extra for a nice data plan on the cell phone. Willing to pay extra because then you get YMAX or LTE, even faster wireless internet on the go. So that might be something that we really need to see those things in comparison. Now, how does that help us with safety, with congestion and with the environment? Well, it can warn the driver, again, there's something coming up, but it can also keep the driver awake. It can also distract the driver for sure. We need to talk about this, but it might be something that the driver has a companion on the road and therefore also that might make things a little bit safer. But also, obviously inform the driver about the best travel option, therefore mitigate congestion because this one motorway might be closed, the other one might be open. Not necessarily revolutionary ideas, but with the power of the internet where everything is connected with everything, bringing this to the car should help with many of these problems. And then also to educate the driver to drive more efficiently in terms of to maybe give the driver recommendations depending on the specific road segment where you might get a lot of data from other vehicles who drove the same road segment and you might wanna say you might wanna be a little bit easy on the right foot because there's congestion coming up anyway, so it doesn't make sense to be speeding right now. So how do we do this and what are the specific questions that we need to answer in that regard? One thing is if we connect all these vehicles and how can we really improve the safety and efficiency by sharing all this data because we need to share the right data and also we need to agree with all car manufacturers more or less that we share exactly this data and that everyone is okay with this. So it's a huge standardization effort to do these things. And then if we say keep the driver awake or improve productivity that you do get, some of your work done while you're commuting to work that it's actually not distracting. And that is certainly something the internet is already in the automobile or we have to do it right and safely. And how can we benefit from existing communication infrastructure? What I said just a minute ago, the infrastructure cost to put in those dedicated communication systems can be very, very high. So can we actually use maybe the cell phone network for these things? And then crowdsourcing. Basically everyone is talking to everyone. Many people go through the same situation but don't really share this experience. How can we use this? What we basically do for social networks and these sorts of things. How can we apply this for a safer and more efficient traffic scenario? And then again the integration with other mobility options which would certainly help us with a more efficient and hopefully also more timely traffic. So again, the internet is in the car. Everyone brings a relatively smart phone to the car these days but how do we integrate this to the car? How do we do this in a safe way? How do we do this like a real integration so that also some data from the car can be used on the carrier? Let's take a look. These are all the different things that the car knows about itself and about its environment. I'm not going to read all of this. There's not going to be a test on any of this. But just browse through and you will see things that you know about. Other things will say, really the car knows all of this? Yes, the car knows about other vehicles that are nearby if you have an adaptive cruise control. The car knows exactly on which seat someone is sitting. Actually how much you weigh and if you are sitting appropriately or lean forward or backwards, all of these things. It knows about the weather more or less about ambient pressure and temperature and all of these things. But the car's a very introverted thing. It keeps all of this to itself. For a number of reasons. For one, because it needs actually the pressure and temperature information to control the engine. Why should it share this information with anyone else? But also the automotive field is obviously a very, very safety critical field and therefore sharing this information comes with certain challenges. So the question really is, and this is where we are actually doing quite some work now. What of this data can we actually use and share among vehicles? Share with an infrastructure so that we can actually use crowdsourcing like every vehicle is maybe sending out information about the dynamic systems of ABS stability control. As you're going through a turn on a particular morning, the ABS kicks in earlier than it typically does. Well, one other vehicles. Already at 0.3G lateral force, the ABS kicks in make this information available. What about driving distraction? I'm not putting on any particular good or bad scenario of a dashboard. It's an Audi A8. I think it's a fantastic car. There are already a lot of things in there. And if you read all the different reviews, you will see certainly that it's a great car. Every once in a while, you'll read that things might be distracting, especially if a cell phone is being added to this. But what does distraction really mean? To this cockpit, it looks actually a little bit more confusing to me. But again, highly trained professionals doing this. And one of our professors, Professor Cliff Ness, who's actually doing all this research, like how do people interact with an autonomous car? And he also researches driving distraction. They looked at this thing. What actually happened when this plane landed in the Hudson River, I think now about two years ago, wasn't this like multitasking that supposedly does not work and then way too much information? Well, no, because everything was directed to one thing. Save the plane. And then people can deal with most of these things. We sadly also saw with the Air France flight, the crash between Rio de Janeiro and Paris about two and a half years ago, I guess, that if the automation is switched off suddenly and then the pilots are exposed to a situation they are not used to, that there are a lot of challenges. Well, again, the intersection of some sort of autonomy, automation and the human being is very challenging. So it has to be done right and it should not be something where the driver then brings in an additional gadget that doesn't really quite fit in there. So I have the thinking, what if you actually have something like a nice docking station that we basically have for all laptops? This thing has a docking station at Stanford, slide it in, I have a nice larger screen and a keyboard and everything, it just works nicely. What about having the same for cell phones in the car where you're using actually the controls of the car to navigate the systems of your cell phone? Should not look like this? Did you have your nice little Facebook on where you fiddle around with something? But what about this? Just three options. Post your location, listen to a post from all your friends or just exit and that's it. Not like, ooh, I'm really excited to go to this cafe and everyone should go there and the latte is a little bit pricey but worth it. Not sure if that's so important to me, but just those three things. And this is really where we also need common standards. I'm very surprised there's some research now going on where it says if a task takes more than 15 seconds to do while driving, then it's considered dangerous. These 15 seconds I find an interesting notion. It's not 15 continuous seconds, but it's the 15 seconds. And therefore in 15 seconds, obviously many things can go wrong. And so again, a lot of research that we have on this field driver distraction and also what the interface should look like. Which infrastructure to use? I wanna be brief about this one in the interest of time. In Europe, I understand this is being called car to car communication consortium where the cars talk to one another, talk to an infrastructure with the benefit or with the motivation to make traffic safer and more efficient. In the United States, we've done quite some work under many different names. My other thing is the current name is Safety Pilot. It used to be Intellidrive. Some sorts of DSRC, Dedicated Short Range Communications are in there. Basically what you say, what if cars could talk to one another? But the question is, since the top row actually uses this proprietary network, only for cars, therefore safer, therefore not so easy to hack, therefore the network load can be balanced more easily, but it costs so much. And there has been a lot of talk about those topics in the United States for about the last 10 years now. Safety Pilot, you see all these test cases that they are doing, but it's not really getting off the ground. And now in about two years from now at the end of 2013, there's going to be the decision about the deployment scenario for Safety Pilot. A topic that has been worked on or will have been worked on by then for almost 15 years and then there will be a decision when they will roll it out. I have to say I'm not the absolute expert in that field, but the experts who are working on this, who are in all these conversations with the Department of Transportation, they say we will not have this before the end of this decade. So even if there's a decision in two years from now, this will take at least another seven years. And then we saw those numbers earlier, how many lives are being perished every year. The question is what can we do with the evolution of cell phone technology? It's not as good as a dedicated system, this is for sure. But this is where I say it might not be safety critical conversation or communication, but maybe safety relevant. So I would not do something like I'm here, and then ooh, this SMS got lost or the call didn't go through. But maybe something, a car go through a turn, what I mentioned earlier, and it's icy in one particular morning, and you send this data to a server and every oncoming car pulls data from the server and knows yeah, there's something in this turn you might wanna slow down. Can save some lives, I guess. And it's actually the same technology that we use for some certain things. I walk up with my cell phone into a bar. Ooh, there are interesting people in this bar here today. So we can probably use much of this technology to make vehicles safer. And this is where we want to start a study actually now early 2012 to look what we can do with existing or soon to be existing LTE, for instance, communication technology to make vehicles safer. Of course, we will need to talk about latency, we will need to talk about availability, but you have to start somewhere. Again, crowdsourcing, I think we talked already about some of these things. And maybe quite some of you might know this picture because this picture has been shopped around, I think, for at least five years now. There are a lot of BMWs in there, but I know that the red three series convertible down there is the E36 for those of you who are familiar with those numbers. And the E36 was, I think, discontinued in 98. So great ideas, same as the book that I looked at for my parents, the future, but how do we do this? How can we really deploy all these things? Now, what about extending the mobility? So this is the part where I said earlier, what if we can do some intermodal connections between the automobile, the personal mobility and maybe other fields? And yes, I updated the slide, particularly for you. I hope that I'm getting the numbers sort of right. So I'm saying, well, if you get information, well, to get to your destination because traffic is not that great, might take you 45 minutes. But there's a guarantee that if you go to the park and ride place, that you'll be there in 32 minutes. Guarantee, that's a tough thing. We might have another researcher working on the legal implications of this, but lots of information is available. So we need to think about how can we use this in a truly automotive or mobility applicable fashion so that actually people use these things? And also that we need to be realistic. Use these things, do we expect a revolution? All of a sudden everyone is doing this and nothing else. Do we say, well, if it starts slowly, hopefully the development cycles and the attention span of the IT industry and the mobility sector play out somehow, not that a certain field might say, ooh, we did this for two years, now we just close our website. Or the other thing that you say, you know what, there's a couple opportunity. There's someone waiting actually along your route and likes the same music that you like, is interested in the same business that you like and actually has the same destination that you have and gets you free beer tonight. Not advocating drinking, beer and driving, but have to come with an example somehow. So if you had all this information and you certainly know that many of these things are just available, cell phone knows where you are, cell phone basically knows who you are and knows who you would like to get in touch with, so make it available in the car. I think that we can do a lot of things and this is where true to the model bringing automotive and Silicon Valley together where we have maybe not so much research but a lot of student interest in that field where they just say, hey, we want to develop this iPhone app and use it in the car. So in summary for these two parts, autonomous and connected, is it actually smart or not? For autonomous driving, I truly believe the technology would be there. We could do this if we said as of January 1st, 2000 and say 2014 in about two years from now, only autonomous cars are allowed, not just on this one road from Princeton to New York which some people want to advocate, but everywhere, we could probably do this. But at what cost? That's the question. And same thing, having vehicles talk to one another, that is possible, but it's a little bit like who buys the first fax machine? You get the first fax machine and then what? Well, if there's another person and that's the person you're doing business with uses the fax machine, it's a huge advantage because you have a huge advantage above everyone else in the same industry. But who buys the first car that can talk to another car? Okay, let me do this. And then someone else buys the car. But I might never literally run into this other person. So it really needs a huge market penetration so that vehicle communication actually works. So I'm talking vehicle to vehicle for safety applications. So really the challenges are the mix of standard and autonomous vehicles. The transition from today to tomorrow, there will be a huge one. And then infrastructure costs are just gonna be immense for vehicle communication. But still, people are asking me, get out your crystal ball and what do you see? So maybe by 2030, we might see completely autonomous cars on some limited roads. You might get other statements from people in the industry and academia. They say autonomous driving by 2020 or something like this. But ask them what they actually mean. They might mean autonomous parking. Did you basically get off the car in front of your restaurant and then your car basically pulls it safe into the side street? It's possible. That's nice. That's very challenging. But maybe not exactly what you expect from this picture they showed you from the early 70s. Vehicle communication, maybe 2020, what I told you if the Department of Transportation makes this call in two years from now. But maybe with cellphone technology already much earlier. So far about the two topics on the left. I don't wanna spend maybe let's say five minutes, I guess in the interest of time, not sure how we are doing. Five. Okay, yes. Then I want to talk about the alternative energy a little bit and with a big topic of electric vehicles. What I say, this is a talk that I gave earlier this year to people who were visiting from Germany or university and they were certainly very interested in electric vehicles as well. And we had a very good discussion about these things and more than happy to discuss this later on here as well. So are electric vehicles really the next big thing? Probably I should have said electric mobility and there I wanna be clear what by my understanding or to my understanding the differentiation between vehicle and mobility is. A vehicle as such is nothing. It's the person who's using it. If the person is using it, it's putting 15,000 kilometers on the odometer every year. Might be dangerous. Might be an obstacle, might be congestion. Might be joy. Might be just a means of transportation. You really need to look at both. The car and the person who's using it. So that's why I'm talking about mobility. So the slide that we talk, no, that's the next slide. Sorry, first the key questions. The key questions that we need to answer and at one point I was actually reframing this slide like which question are we actually trying to answer? Was electric mobility? And that's not an easy one because very often then I read publications that say, oh, electric vehicles are more expensive than conventional vehicles. That's not a surprise I find. And then they do a lot of these calculations. How many miles you need to drive so that it might break even and all of these things. Well, yes, it might be more expensive. But the question to my understanding was not, what is the cheapest way to get from A to B? We sort of figured that out what the cheapest way is. But the real question is, can this help to slow down global warming? My one first versions I had prevent global warming, but unfortunately I think this is the right phrasing for it. Decrease the dependence on foreign or scarce resources. And this is obviously something that's also very interesting to look at the discussion in different countries. In the United States you talk a lot about foreign resource and independence from foreign oil. In other parts of the world you just listen more or get more the notion of we are running out of oil so we have to find a solution. And decrease air pollution. Talked about pollution already. And then also what we discussed quite a bit and what I guess what ESB is heavily working on. How does it integrate into the whole electric infrastructure? If all of a sudden everyone buys an electric car, can the grid handle this? This is what we need to look at. But maybe also where the benefits, if we all of a sudden have all these batteries that might act as a buffer on the grid. And then actually become the better alternative for consumers. Consumers say yes, I want one of these cars. Remember, it's not just about technology and the consumer wants something. Is it maybe something that needs to be incentivized so that people subscribe to this technology? And there's probably not just one answer. That is the important part in this. So to give that away already, the answer to all of this is electricity or electric mobility, yes. The question is how much? Talking about how much. This is a slide that you see over and over again in presentation in the United States. How much do people drive? What you see on the vertical, sorry, on the horizontal axis is the total travel distance per vehicle. It's not per person, so it's a little bit different per person or per vehicle. But it's basically how much does one vehicle drive per day? And on the vertical axis, you see the cumulative daily distance, which is basically what percentage of vehicles drive exactly this range. So you see the line on the red line on the left is basically that 50% of all vehicles drive less than 20 miles every day. And actually 80% of all vehicles drive less than slightly less than 45 miles a day. For the answers, everyone should be happy with 100 mile range electric vehicles. 160 kilometers, as you certainly know. And then the question is, well, everyone should be happy with those electric vehicles that we have today. Because the cars fit in very nicely there. And I really appreciate the Nissan Leaf a lot. I think it's a great step. And we also had Carlos Ghosn to Stanford University and told us a lot about why they are putting so many resources behind this and why they believe in this. And his answer, Carlos Ghosn's answer, is actually not, well, because of global warming or because of pollution, all of these things. He says, these are all very important things, but the society expects it from us. So that is really something. The society expects it from the corporation. Again, there you have this tension or this balance between consumers, corporations, policies, and so on. So that's the electric vehicle fit in there. You could say, of course it fits in there very nicely. But the reasoning when buying a car is a little bit different because you don't look at it like, okay, I know that I'm basically driving my 30 kilometers to work and back every day. It's basically, what am I, there we are. What am I gonna do over the next five years that I'm going to own my car? So you think at all of these things and you think, will this car be able to get this job done? And very often the answer is, I'm not sure. And therefore it's a relatively safe bet to say, ooh, my next car is not going to be electric, maybe the one after, but for now I stick with conventional technologies. And this is what we have to understand. As an engineer, I could say, solve the problem. Actually 90% of all people should be totally happy with 100 mile range. But in fact, people, people means all of us are looking at the topic very, very differently. And this is where we are now doing a lot of research actually, what is the motivation and maybe also the demotivation for people to look at a certain vehicle. So this is basically, I want to be brief with this, a questionnaire that I got from people in class. On the top, if I ask the question, what is going to be your next car equipped with a powertrain? About 60% say internal combustion engine and only 20% and this is Silicon Valley students say, electric. But if I ask the question, there's a car you can test next weekend and you can try it out. Which car do you want? More than three quarters, one electric car. So there's a lot of curiosity about this. People want to try it out, but they don't want to commit for the long term. This is what we need to understand and this is where there might be a connection with car sharing and things like this. Here's another one, I want to be brief about this, comparison from the United States and Europe again. Very, very similar, the absolute numbers differ quite a bit, but the top in the United States, consumers were asked, what are the challenges they see about electric vehicles? And I guess it's no surprise, range and battery life and the charging infrastructure. And the second row is an earnest and young report from August this year where they asked, not consumers but industry experts and it's the same order where people just say, the range is a problem and the charging infrastructure. So the question in the end is, are consumers ready for electric vehicles or electric vehicles ready for consumers? And that's also not necessarily the same thing. And one last thing I wanted to mention here before maybe we go to questions, looking again at this slide, and this is in a way my realization of the year, what's actually pretty interesting that we're spending so many R&D resources and focus on the right part of this chart. But I'm not sure if it's really the correct, the right part of this chart, because if we are trying to solve this piece, basically what happens after the 40 miles and put even bigger batteries in the cars, more powerful, more energy in the car, gets heavier. And we are designing a car for a purpose where it's not being used that often. We end up with a pretty heavy vehicle, but it's not being used for this. And that sounds, which I found very surprising, very much like an SUV. Or a car with all wheel drive because you might go to the mountains in winter. So we need to be clear about this that most of the driving happens here. And therefore, our vehicle option might be different. And this is why a car that can travel most of your driving in an electric mode but gives you the peace of mind that you can do all these other things, maybe with a combustion engine that is lighter than a heavier battery might be a good solution. There needs to be much, much more work to be done in this field, but we also have to be careful that we don't just look at the right part of this chart and say, better batteries, bigger batteries, more range, we might need to be more realistic. If everyone is just driving a little every day, let's put in a small battery. We need charging infrastructure in all of these things, but we have to be realistic. People want to do all of this on the right and therefore, what is the best option to do this? And this is where we get into life cycle assessment where you can see, I guess we're gonna make these slides available afterwards so you can look up the reference down there and the source. It's a great study from our dear colleagues at Carnegie Mellon University where basically they looked at the life cycle emissions, damages and premium for oil over the lifetime for the CV, which is a conventional vehicle with an internal combustion engine, the hybrid electric vehicle, the plug-in hybrid electric vehicle was 20 kilometer range, was 60 kilometer range, and the battery electric vehicle was 240 kilometer range. And they looked at different scenarios on the very right. The base electric mix in the United States was certainly a lot of coal, but also an optimistic case where you have all electricity generated from zero emission power plants, obviously renewable. And but what if you have everything coal-fired power plants where you then see especially the production of very heavy and very high energy containing batteries is just generating, as you can see, a lot of sulfur dioxide, which brings us to the piece of pollution, climate and resources again. So we can now break down these things. Greenhouse gas emissions certainly is in there, but it's not the only question to answer. We need to look at the resources and we need to look at the pollution part of it. So in conclusion, we do have many of these options already today. We have relatively clean conventional vehicles, diesel vehicles that come with challenges of their own. We have microhybrids, we have mild hybrids, we have full hybrids, we have plug-in hybrids, we have range extended battery electric vehicles, we have battery electric vehicles, we have battery electric vehicles with an incredible range and on the horizon might be fuel cell vehicles. But the question really is, and this is where the crystal ball is again, which is actually the right vehicle that we can put into these segments, is the technology ready? Is the consumer going to subscribe? Is it a viable business case for cooperation? Does it fit in the political landscape and does it help the environment and the resources? With this, we need to see the future of the car, I believe is electric, but I'm not quite sure how much and this is where we will spend more research. I hope there was informative to you so far and I'm looking forward to discuss questions with you. Thank you. Thank you.