Veel gestelde vragen over de WEpods

Vragen over technologie:

De 3 meest belangrijke redenen waarom automatisch rijden belangrijk zijn:

  • Verkeer stroomt beter door;
  • Minder ongelukken;
  • Minder brandstof nodig.

Verkeer stroomt beter door
Als auto’s met elkaar communiceren, kunnen ze remmen en optrekken beter op elkaar afstemmen. Zelfrijdende auto’s reageren eerder wanneer voorgangers remmen. Zij houden ook automatisch een bepaalde afstand tot de auto ervoor aan. Hierdoor kan bijvoorbeeld minder snel een file ontstaan.

Minder ongelukken
Menselijke fouten zijn meestal de oorzaak van verkeersongelukken (in 90% van de gevallen). Met zelfrijdende auto’s is er daarom naar verwachting minder kans op verkeersongelukken. Dat is beter voor de verkeersveiligheid.

Minder brandstof nodig
Zelfrijdende auto’s rijden vaak zuiniger. Vrachtauto’s kunnen bijvoorbeeld dichter op elkaar rijden. Hierdoor hebben ze minder last van luchtweerstand. Auto’s die met elkaar communiceren hoeven minder hard en minder vaak te remmen en op te trekken.

De voertuigen hebben op hun rit tussen Ede en Wageningen een hospitality functie voor gasten. Daarom worden ze uitgerust met een groot scherm waarop informatie over FoodValley wordt gegeven. De informatie is plaats bepaald en laat bijvoorbeeld zien wat er in een gebouw waar het voertuig langsrijdt.

De WEpods kunnen 40 km/uur rijden. Tijdens de test rijden ze niet harder dan 25 km/uur. Ook na de tests zullen de WEpods 25 km/uur blijven rijden. Hoewel er geen zichtbaar rempedaal is, zit er uiteraard wel een rem in de WEpods.

Verschillende boordcomputers combineren gegevens en geven opdracht aan onder meer de remmen en het stuur. Op deze manier kan het voertuig goed anticiperen op plotselinge veranderingen. Het voertuig kan dus snel en adequaat reageren. Het heeft wel, net als elk ander voertuig, een remweg.

Voor de WEpods wordt een sensorsysteem ontwikkeld dat de omgeving in beeld brengt om:

  • Andere verkeersdeelnemers te zien, te classificeren, te volgen, hun snelheid te meten en hun route te voorspellen.
  • De navigatie te ondersteunen door middel van ‘landmark positioning’; het meten van de voertuigpositie ten opzichte van kenmerkende vaste objecten langs de route.

Om dit te realiseren wordt het voertuig rondom voorzien van redundante sensoren die werken op verschillende fysische principes:

– Camera’s rondom
– Radarsensoren rondom
– Lasersensoren op de vier hoeken

De data van de omgeving sensoren zullen worden geselecteerd en gecombineerd (sensor fusion) voor een robuuste schatting van de positie en snelheid van relevante objecten in de omgeving.

Navigatie betekent dat het voertuig zijn weg kent en in staat is die veilig en betrouwbaar te kunnen volgen. Hiervoor wordt een speciale, zeer nauwkeurige landkaart van de route gemaakt waarop naast de geometrie van de weg ook alle zichtobjecten (bomen, lantaarnpalen) zijn opgenomen. Daarop wordt de precieze bewegingslijn van het voertuig geprojecteerd en wordt de (maximum) snelheid per wegdeel vastgelegd. Het maken van een dergelijke kaart is nieuw en de exacte definitie moet door het project worden ontwikkeld. Het voertuig moet heel precies en heel betrouwbaar zijn positie op de weg en op de kaart kunnen bepalen.

Daartoe wordt het voertuig uitgerust met een aantal redundante systemen:

Zeer nauwkeurige GPS sensor voor satellietnavigatie
Inertial Navigation sensor om in alle richtingen de versnelling van het voertuig te meten
Odometrie: Sensoren om wielomwentelingen en stuurhoeken te meten en zo de afgelegde weg te bepalen
Camera om, indien nodig, lijnen te volgen
Radar (in combinatie met camera) om de exacte positie ten opzichte van vaste objecten (landmarks) te meten.
Het voertuig zal altijd door een combinatie van deze subsystemen zijn positie bepalen.

 

Het voertuig
Als we het hebben over de WEpod, dan hebben we het over de EZ-10 van EasyMile uit Frankrijk. Dit elektrisch aangedreven voertuig heeft veel voorzieningen voor automatisch rijden. Het WEpods projectteam heeft vervolgens technieken toegevoegd om het voertuig geschikt te maken voor rijden op de openbare weg.

De belangrijkste kenmerken van de EZ10 zijn:

• 6 persoonscabine, met automatische deur en uitschuifplateau voor rolstoelgebruik. In plaats van de rolstoel kunnen ook 6 personen staan. Voor het project gaan we uit van zittende passagiers.

• Hoogte 275 cm, breedte 199 cm, lengte 393 cm, draaicirkel 5 meter

• Maximum snelheid van 40 km/uur (tijdens de proef niet harder dan 25 km/uur)
• Elektrische aandrijving met een bereik van circa 100 km
• Opladen d.m.v. een conventionele connector (handmatige verbinding)
• Eenvoudige verwarming en airconditioning (in verband met energiegebruik)

Door: Prof.Dr.Ir. I.R .van de Poel

1. Describe the case (or design project) you are working on in thispracticum.

A new technology that is recently in developmentis the Google Driverless Car. They aim is to get this car available on themarket within five years from now (Priddle, 2015). Google is engineering andbuilding driverless cars in Livonia in partnership with metro Detroit suppliersin a marriage of California tech and Motown know-how (Priddle, 2015). Thesecars have to be able to drive without any intervention of a human driver butcan also be driven manually.

Driverless cars are recently allowed by lawin many states in the United States. Some states like Georgia have decided notto change its laws to allow autonomous driving. The car will be tested in theNetherlands by non-experts (regular drivers) in 2015 and it is expected thatGoogle Cars will drive on the Dutch roads in 2020 (Priddle, 2015).

This is the case that we choose to base ourVSD envisioning cards on. It is an interesting new technology because of thechange in responsibility from the driver to the machine. There is a lotdiscussion about this subject and this impedes the implementation of it.However the technology could provide a lot positive effects in the field ofhuman well-being, safety and sustainability. Yet, there is a lot of uncertaintyabout how the technology will contribute to these subject matters. This makesthis technology very controversial and thus interesting to apply the VSD cardson.

2. What card(s) did you choose and why? Please repeat the assignment on the card(s)!

2.1 Stakeholders: Non-targeted Use
We started with the cards for thestakeholders, because it is important to first consider all involved partiesupon further investigation. Within the category ‘Stakeholders’ we decided to take the card for non-targeteduse. At the point of non-targeted use the most dangerous situations can occur. The assignment says: “Technologies arenot always used as in ways that the designers intended. Who might use thesystem for unplanned or nefarious purposes (e.g. frustrated stakeholders or anidentity thief)? In what ways? Identifythree roles that involve non-intended use of the system.”
2.2 Time: Long-Term Health and Well-Being

For the second assignment we chose a cardfrom ‘Time’. The card we picked is Long-Term Health and Well-Being. The card wepicked is: “Technology may have bothdirect and indirect effects on people’s health and well-being. How mightinteractions with the system on a daily basis influence health and well-being? Imagine that the system you are working onhas been widely adopted and is part of daily life for direct and indirectstakeholders across society. Reflect upon 3-5 likely ways in which the systeminfluences health and well-being after years of use.”

Some group members associate the Google Carwith the movie Wall-E, in the movie also futuristic cars are used and you seethat people become lazy. In the movie Wall-E people are watching television andthey all become very fat. This is a bit exaggerated but indicates that peoplewill become lazy.

2.3 Values: Choose Desired Values

Assignment: Deliberate on and then designate three primary values the system ideally would support. We’ve chosen this card since we would say it is important to see whether it is feasible to meet the desired values. The values we consider to play an important role for the Google Car are safety, freedom and sustainability.

2.4 Pervasiveness: Crossing National Boundaries

Assignment: Choose three countries across the globe and envision challenges for your system if it was deployed in each of those countries. Label any common concerns across the identified challenges. We picked this card since the geographic location might affect the successful implementation of the Google Car. With the Google Car you might want to cross borders for example when you go on holidays.

3. What are the main outcomes/results/conclusions of the assignment?

3.1 Stakeholders: Non-targeted Use

Criminal intentions

Cars are important during criminal activities. Cars are important for bank robbers, drive by shootings and the like, because they are not only the method of escape but are also capable of carrying more load. The advantage of the google car is that it drives autonomously. This means that for example all criminals can be shooting the pursuers instead of paying attention to the road, resulting in greater accuracy. In the near future depending on the development of the technology a google car could be a better “driver” than a human driver because it is capable of reacting faster and does not have to deal with the emotions caused by the criminal activity. The cars could for this purpose also be hacked to make the advantage larger, for example by programming them to ignore traffic lights and speed limits. This combined with the shorter reaction time could vastly increase the potential of getting away, the car could for instance ignore a traffic light and still be able to manoeuvre through the chaos. Another treat is the use of self-driving cars for terrorism. A google car could be filled with explosives and programmed to hit a target with pinpoint accuracy. (Harris, 2014)

Driving under influence

If you have an autonomous driving car the thing you want to do is just relax, read a newspaper, watch a movie or do something else while you are being driven. This is fine, because if the system fails or gives an error, you can take over control yourself at any time. But when you are drunk or under the influence of other drugs/medication, you cannot take over control and drive yourself home safely. And since it is stated by law that there should be the possibility for humans to take over control at any time, you need to be sober to sit behind the wheel of a autonomous driving car at all time.However, in some states of the USA driverless cars are allowed on the road, so there it would be possible to go to a bar, have some drinks and then let yourself being driven home by your google car. (Kelly, 2012)

Driving at night, for holidays

Falling asleep behind the wheel can cause very dangerous situations. To minimise the risk of falling asleep a campaign has been released a few years ago by the government. Billboards were placed next to the highways which tell drivers when to take rest and how to recognize fatigue symptoms. With an autonomous driving car the chance to fall asleep becomes greater. As a driver of an autonomous car you only have to check if the car is driving accurately but you do not actively have to control the car. This makes it more difficult to remain focussed. A good example of a situation when people sooner fall asleep is when you drive to your holiday destination. This happens mainly in the night at times when people usually sleep. Also drivers tend to stay longer behind the wheel than advised to arrive sooner at their destination. There does not necessarily have to be a direct impact if you fall asleep in a Google Car, since the car can continue driving. However, in case anything happens you are not directly capable to respond to this. Driving the Google Car still requires the driver to be awake, capable and in the possession of a driver’s licence to take over the control in any situation. (Rijksoverheid, 2014)

Driving without a backup human driver

Having cars that can drive without a backup can be very handy in certain situations, such as the replacement of truck drivers, or for instance bringing your children to school safely. In the future this might even come true but for now it remains illegal to have an unmanned vehicle in the biggest part of the world. Only the states of Nevada and California (Kelly, 2012) allow the testing of driverless cars, but this is still only in test phase and humans remain inside the car. Cars operating by themselves is not the intended use of the google car right now. They will probably work towards self-driving vehicles as that is a logical continuation on their current work.

3.2 Time: Long-Term Health and Well-Being

Passiveness

It is rather difficult to say what theimpact of mass use of the google car will be on well-being and health. With alot of inventions we see an increase of well-being at first, but after theinventions have been used a while, the negative impacts also become visible. Tounderstand the impact of an autonomous car, like the google car, can have onwellbeing and health, it is important to first look at the impact of two othertrends. These trends are the increasing use of cars and motorcycles and theincreasing use of the internet and social media.

The past hundred or so years, the movementof people went from being primarily walking or using a bicycle to the use ofcars and motorcycles. This step has had a large impact on well-being, bothpositive and negative. Because, even though, people and goods could betransported faster and further away, however people also started using cars andmotorcycles for all distances including tiny distances, resulting in lessphysical workout. The lack of physical workout is not good for the averagehealth of human beings.

Another trend with a major impact is thewidespread use of the internet. People have started to use social mediaconstantly, especially with the presence of smartphones and tablets. Studieshave shown that people visit each other less often and do not speak to eachother face to face, but only via typed words on screen (de Vos, 2013). The constantuse of social media and the internet does have positive aspects on wellbeing,knowledge is spread more easily which causes crimes or illnesses to be detectedand cured easily. Also people can stay in touch with friends they wouldotherwise have lost, which also benefits their wellbeing. But it is importantto know that to stop seeing each other in real life is a very bad substitute,and this could decrease the wellbeing of people.

We will now start with the expectance ofthe influence on wellbeing and health of autonomous cars. It would not bestrange that within 20-30 years from now, all human piloted cars are phased outand even banned. Autonomous cars can have a positive influence on wellbeing andhealth because they can become safer than human driven cars. Autonomous carshave a faster response and can more easily avoid accidents. Another positivething is that transportation will be available for everyone, as no driver isneeded and self-driving taxis can therefore become quite cheap. This means thatowning a car will not be necessary, which has two results, first that it willbe possible for more people to use transportation and secondly that less carshave to be made which is good for the environment. However, as we have seentwice already, there are usually good and bad impacts of an invention or trendon wellbeing and health. Maybe 20-30 years from now, no driver will be neededto watch the road, people will be able to just step out the door and get into aself-driving car. This will cause people to become more passive because theydon’t need to pay attention to their surroundings. This can make time to doother things, which can be both good and bad for their wellbeing. A largepossibility is, however, that they will communicate via social media even more,as is already visible when travelling by train. This, combined with the factthat the person in the car does not have to pay any attention to thesurroundings, can cause people to distance themselves more and more from thereal world.

An example of this can be seen in the movieWall-e, which is set in the future. In this movie, people float around inself-moving chairs and look only at a screen that is in front of them. When thelead of the movie, a robot called Wall-e, causes two people to fall out oftheir chairs they are helpless but do find out how enjoyable it is to have realcontact with other people.

Safety

Thenumber one cause of death for young people is driving accidents. Almost all ofthese accidents are caused by human error. With the Google Car the human erroraspect will be eliminated. It is assumed that accidents can be prevented by theuse of machines. The chance of a machine failure is significantly smaller thana mistake by humans.

We gotthe tip to look at the following disaster scenario: “What if in the early stages the technology worksreally well, and society moves quickly to adopt Google-cars. One day, thehighway filled with Google Cars, a datacenter burns down, and the cars suddenlylose their internet connections and instant chaos results (cars crash into eachother, people are hurt, ambulances can’t get reach them because the cars aretoo densely packed on the road, panic,

etc). After 15 minutes theinternet connection is restored, but the damage is done. What would this do forpublic acceptance of the Google car? What could be the consequences?” (van de Poel, 2014)

We think this situation would be very unlikely sinceevery google car has its own computer and set of indicators monitoring trafficall around them. Thus to create a huge disaster all google vehicles on ahighway will have to breakdown simultaneously which would be highly unlikely.Data centers would only be necessary for route determination and to not knowthe route for fifteen minutes or any amount of time wouldn’t matter as thegoogle car would just stay on the highway and drive.

Time efficiency
The GoogleCar can save time in several ways. An interesting one is the fact that the sizeof traffic jam can be diminished by the use of a driverless car. The capacityof the highways can be increased by a factor 2 or 3, since cars can drive muchcloser to each other. The human responsiveness does not have to be taken intoaccount. Probably the number of car users will increase in the future, but thiswill not outweigh the capacity gain with the Google Car.

Another obvious aspect in time efficiency is the fact that drivers can do otherthings instead of continuously being focused on the road. Like in publictransport you can for example work/read things while your car is basicallydriving you around. This also offers opportunities to redesign the car in such away that it might look more like a small replaceable office.

Environment

Since the environment is very important inour existence we deem it important for well-being of humans and thus importantin this assignment.  We think that thereare two scenarios for the impact of the google car on the environment, ashort-term impact and a long-term impact.

In the short-term, let’s say the first fiveto ten years, when all regular cars will make place for self-driving vehicles.There will probably be a lot of vehicles that are not really at the end oftheir life but they will be traded in and wrecked. All this extra material thatneeds to be recycled or landfilled will have a great impact on our environment.The recyclables will consume a lot of energy to be recycled but they can beused for building new google cars. The non-recyclables will be landfilled orincinerated, both are really harmful for the environment (Yang, 2014).

In the long-term, let’s say in about 30years, when all regular cars are gone and everybody owns a self-driving car.CO2 emission per car will go down since the computerized car will drive muchmore efficient than a human driven car. So this is a really good thing for theenvironment.

The google car itself is a lot smaller thanmost cars and thus materials could be spared if all car branches producesmaller cars, but this is not very likely to happen since there will always bepeople that want to have big cars to show off. The new computerized cars willall need pretty big computers to process traffic data and all surroundings. Thematerials needed to build computers will become scarce or critical materialsand thus we will probably delve deeper and deeper into natural resources, whichis very bad for the planet.  Thematerials used in processors and other computer parts are usually rather hardto recycle and thus recycling will probably become very energy intensive.Energy that has to come from somewhere and will probably contribute to globalwarming thus will not be very environment friendly.

3.3 Values: Choose Desired Values

Safety

With a product like the google car it is ofutmost important that it is safer than a human driven version. It mustabsolutely be avoided that a google car is the cause of an accident. Accordingto Ibo van de Poel, people consider risk caused by own actions significantlymore acceptable than risk caused by something/someone else’s action. As the carcan be driverless the risk is only partially caused by the people inside thecar. The people inside the car will not feel and are not responsible for thedecisions the car makes, they are however responsible for using the car (ICAO, 2008).If a google car does cause an accident it could have a negative impact on thepublic opinion concerning autonomous vehicles, possibly resulting in lawsuitsand protests against the use of autonomous vehicles, hindering its development.

Freedom
One of the best values of the google car isthat you have the freedom to do whatever you want when you are beingtransported. This will also improve human well-being. You do not have to payattention to the road or have the feeling that you are wasting time when youare in a traffic jam or need to drive for 2 or 3 hours to your destination. Itis now possible to relax, watch a movie, read the newspaper, do your work orhomework or even to just sleep while you are being transported. And to do allof this, you also do not need a driver’s license because you yourself are notdriving the vehicle. This means that everybody has the freedom to go whereverthey want whenever they want. Many people will think this is very good and muchmore time efficient for a lot of people, but it also has a downside. Almosteveryone will want to have such a car, which means that there will be much morecars on the road and we would need much more parking spaces.

When you are talking about amount of carson the road, you do need to consider that autonomous driving vehicles will beable to drive much more closely to each other and therefore this does not meanthat there will be more traffic jams. However, studies show that in 2013 we hadalmost 1 car per 2 persons. This means that if everybody would want a car, theamount of cars only in the Netherlands would double. A consequence from thiswould be that the emissions would also double and that really is very bad forhuman health. This is a negative scenario considering sustainability.

Now we have talked about the space thatthese cars will take in when they are driving, but all of these cars need to beparked somewhere when they are not driving. In the Netherlands there already isa shortcoming on parking spaces in and around the big cities. If the amount ofcars will double, it would be impossible to park all the cars in and around theplaces where people live. This has a negative effect on human well-being.

Coming back to what I said in thebeginning, for some people a google car will be a very good investment, becausethey can work while they are being driven and this saves them a lot of time.But today there are a lot of students with driver licenses that also cantransport people to their work. When you are looking from this perspective, thegoogle car would be superfluous and it would take away a lot of work forstudents and other professional chauffeurs, not even mentioning taxi drivers.However, some people will have difficulties with the acceptance of being drivenby a robot (ICAO, 2008).This makes us consider to keep the possibility to maintain the freedom to steerthe car.

Sustainability

Sustainability is in our opinion anotherfundamental value that is supported by the Google Car.  Nowadays, the energy consumption is for 35%at the expense of transport (MacKay, 2009). In particular the traffic jamsprovide a large energy use in the form of fuel combustion. Traffic congestionslead to an increase of fuel consumption of the order of 80% (M. Treiber, 2007).A direct consequence of fuel combustion is the emission of greenhouse gasses.This makes traffic one of the largest contributors of the CO2 emissions(Chapman, 2007).

The Google Car is capable of reducingtraffic congestions by applying computer models to the navigation of the cars(A Aoun, 2013). The reduction of traffic congestions will lead to a reductionof the fuel/energy consumption and also to a reduction of the emission ofgreenhouse gasses. The more Google Cars get accepted, the better you will seethis reduction. Since traffic is such a large contributor to the energyconsumption and greenhouse gas emission, the reduction of traffic jams is muchin favor of a more sustainable world. This makes sustainability a primary valuethat is supported by the Google Car.

3.4 Pervasiveness: Crossing National Boundaries

USA

The United States of America may be one ofthe best countries to introduce the google car. The city centers have widestreets, which are asphalted almost everywhere and are easy to navigate becauseof the straight streets and easy corners. Also all the highways have multiplelanes and are good to drive, so you don’t have to worry that much about fastdriving people and slow driving trucks all the time.

However, the google car most of the time navigates on the roads that are knownto google maps, just like normal GPS systems. If there have been road works ora whole new district is built and you need to go past it, the system willencounter a lot of problems finding its way. Of course, there are a lot ofsensors on the car that monitor roads and register any unusual things on theroad, but it could still mean that you would be able to drive yourself throughit more easily than the google car.

Also, if the roads aren’t that good, around farms or in the desert, the carcould say that it is not possible to reach the destination. This could have tworeasons; the road doesn’t exist on maps so the car says you can’t reach it. Orthe road isn’t smooth enough according to the sensors, so it says that thereare obstacles that can’t be driven over or around.

But, a few days ago, a self-driving car from Audi has travelled about 900kilometers in the United States. It drove from Silicon Valley to Las Vegas,fully automatic. This means that it is possible to have self-driving cars inthe United States, but I think they didn’t take the scenic route to see if italso works on small gravel roads.

India

There are a number of challenges that haveto be solved when thinking of introducing the Google car in India. Thesituation on the road in India differs substantially from the situation in thewest, and this poses multiple challenges to different aspects of the googlecar.

The first challenge is the way Google cardetermines its routes and the location of crossings or traffic lights. Whilethe Google car uses many sensors to determine the obstacles around it, it usesmaps to know the route and the location of traffic lights. The consequence ofthis is that the maps have to be up to date for the car to be able to drivearound. If a traffic light has been added, the Google car will not know this,and its camera will not be active to check the color of the light. This canpose dangerous situations, especially in a developing country where a lotchanges in the road network.

The second challenge is the fact that theroads in India are crowded, this means that a large number of objects have tobe registered by different sensors. Seen as the data from these sensorsnormally is about 1 GB per second, one can imagine that with the business ofthe roads in India, the amount of data will only increase. The system might notbe able to keep up with this and the obstacles will not be registered in timeand accidents cannot be avoided.

The third challenge the Google car faces isthat the quality of the roads is dissimilar to the roads in western countries.There are a lot of holes and other unpredictable phenomena when driving there.Google car could miss these holes in the road, just like it is not able to seeopen manholes.

The last challenge is that the trafficrules and mentality on the road is totally different in India than in a lot ofwestern countries. Rules in India are different, for example, the right of wayrules are based on might is right, in which the bigger vehicle will have rightof way just because it is bigger. Another example of different rules is thefact that the indicator lights are used to say someone can pass, instead ofindicating where the car is going. The whole mentality and idea of how the roadshould be used is very different, so the mechanism and algorithms that are usedto maneuver around, will have to change a lot when these cars will be madeready to use in India.

Allin all, much has to be done before the autonomous car could drive in India. Themaps of the roads should be updated regularly in order to make determining thelocation and route possible and the possibility of the system to work with thebusiness of the road should be determined. Also, the car should be able torecognize holes in the road and other unpredictable things, and the systemshould be altered to work with the different rules and mentality in India.

North-Korea

The google car uses a combination of sensors and software to locate itself combinedwith highly accurate digital maps. A GPS is used, just like the satellitenavigation systems in most cars, to get a rough location of the car, at whichpoint radar, lasers and cameras take over to monitor the world around the car,360-degrees. The software can recognise objects, people, cars, road marking,signs and traffic lights, obeying traffic rules and allowing for multipleunpredictable hazards, including cyclists. It can even detect road works andsafely navigate around them. (The Guardian)

In a country like North-Korea however highly accurate digital maps cannot bemade available due to their privacy policies. Currently there is access to somebasic maps from North-Korea. In January 2013 Google Maps published a moredetailed map of North-Korea. Where in the past North-Korea only appeared as agrey area. (De Volkskrant) Figure 1 illustrates the former and current view ofNorth-Korea. However, Street view is still not allowed in North-Korea and theaccessibility to internet within North-Korea is limited.

The google car will not be able to operate properly in North-Korea, because forthe good performance it is essential to have access to detailed maps of theroads. In addition, the cameras installed on the google car are not allowedeither in North-Korea for the same privacy reasons.

4. What possible implications for your case can you foresee, based on your outcomes/results/ conclusions?

4.1 Stakeholders: Non-targeted Use
Implications that can occur with the GoogleCar will mainly take place when the car is driven by distracted drivers,drivers with a criminal intention, drivers under influence or no driver at all.

Since the autonomous cars are not (yet)100% reliable, there still has to be an attentive person that can take over thecar in any situation. When the cars are reliable enough to be driven withoutany backup human driver, it could be possible that this would be allowed by thelaw as well. In this case the mentioned ‘unintended’ use as a result ofdistracted drivers or influenced drivers would be legal and will not belong tothe unintended use anymore. It could even become possible (and legal) to transportloads without a human driver on board. The only unintended use that remains,following from our outcomes, is the (mis)use of the cars in criminal actions.As mentioned before, these cars can probably also be hacked. Criminals can dothis to take advantage of situations in which they are involved or they can usea Google Car as a driving explosive which can be programmed to hit any targetthey want.

The overall conclusion we made is thatthere will be no unintended use with drivers without any criminal intention when the cars are reliable enough to drive completely without a backup humandriver and that this is allowed by the law. The only misuse that remains is theuse of Google Cars in criminal activities. This risk can be reduced by makingit harder to hack the cars. Our opinion however is that this will always bethere. The design conclusion is that the risk for unintended use can be reducedby making the car more reliable and harder to hack. In that case, roads willbecome much safer which is a desirable thing.

4.2 Time: Long-Term Health and Well-Being

What we think that will happen when thegoogle car is widely adopted, is that people become lazy and fat. This because they can be transported wherever they want whenever they want and they do not needto be able to drive themselves, so they stop paying attention to the rest ofthe world and to their health.

On the other hand, driving will become more safely because computers canforesee accidents more easy than people. Also people will be faster on theirdestination and they can do other (important) things while they are on the road.

When we look at what happens to the environment we see that there will be lessemissions because cars can drive more closely together and more economic whiledriven by computers. But there are a lot of rare materials in these cars thatare very hard to recycle, which means that it will cost a lot of energy toretrieve these materials and re-use them in new cars.

Vragen over veiligheid:

De WEpods mogen niet zomaar op de openbare weg! Daarvoor is eerst een aanvraag gedaan (en gehonoreerd) bij de Rijksdienst Wegverkeer (RDW) in het kader van de Algemene Maatregel van Bestuur voor pilots met zelfrijdende voertuigen. Ook daarin heeft de pilot een primeur: WEpods is het eerste project met een aanvraag voor volledig automatisch rijdende voertuigen.

Het Besluit ontheffingverlening exceptionele transporten regelt de ontheffing voor voertuigen die niet in de ‘normale’ categorieën passen. De testaanvrager moet eerst aantonen dat het testen op een veilige manier gebeurt. Daarvoor moet hij een toelatingsaanvraag doen. De RDW kan daarna ontheffing geven.Nadat was aangetoond dat de pilot aan de veiligheidseisen voldoet, is de wet- en regelgeving is hierop aangepast door Rijkswaterstaat. Voor een zelfrijdend voertuig verkocht mag worden aan consumenten, moet de internationale wet- en regelgeving worden aangepast. Dat is nu nog niet gebeurd.

Het project richt zich op de standaard ISO26262 ‘functional safety for road vehicles’ en op kwalitatieve (niet kwantitatieve) onderbouwing van de Safety Case.

De veiligheid staat bij WEpods vanzelfsprekend voorop. WElly en WURby zijn uitgerust met een laserscanner, sensoren, radars en een 3D-camera. Dit allemaal om ervoor te zorgen dat de voertuigen veilig kunnen rijden zonder iets te raken. Het blijft overigens wel zo dat medeweggebruikers natuurlijk ook goed moeten uitkijken, zoals ze altijd moeten in het verkeer. De WEpods hebben immers, net als gewone auto’s, ook een remweg die ze moeten afleggen.

Het voertuig reageert sneller dan een mens kan doen, maar heeft dezelfde remweg als een gewoon voertuig met deze snelheid. Door de sensoren op het voertuig wordt direct een signaal afgegeven en treedt het remsysteem in werking. De operator in de controlroom zal de beslissing nemen of het voertuig na de (nood)stop door kan rijden of niet. Via camera’s kan de operator de situatie beoordelen.

In het geval van onverwachte complicatie in het verkeer, zal het voertuig gelijk stoppen.De WEpod rijdt niet bij een verhoogd risico vanuit de omgeving: in de spits, in het donker, bij gladheid of bij slecht weer.

Vragen over wet & regelgeving:

Het ministerie van Infrastructuur en Milieu heeft dit samen met Rijkswaterstaat en de RDW onderzocht. De RDW is verantwoordelijk voor de toelating van voertuigen op de openbare weg, ook voor zelfrijdende auto’s! De RDW kan zelfrijdende auto’s voor testen toelaten op de weg, maar niet zomaar. Hiervoor moet de testaanvrager een toelatingsprocedure door en aantonen dat deelname aan het verkeer veilig is.

Een zelfrijdend voertuig mag natuurlijk niet zomaar op de openbare weg. Wel heeft het ministerie van Infrastructuur en Milieu grootschalige testen met zelfrijdende auto’s op de openbare weg mogelijk gemaakt. Om een test te mogen uitvoeren is speciale goedkeur nodig van de Rijksdienst voor Wegverkeer (RDW). WEpods kreeg de eerste fase van goedkeur in januari 2016. Daarom ogen we nu testen op de openbare weg, maar alleen op de afgesproken route en onder een aantal voorwaarden. Zo moet er altijd een getrainde steward met rijbewijs aan boord zijn voor supervisie. Om zonder steward of leeg te mogen rijden moet de internationale wet- en regelgeving worden aangepast. Daarover hebben de Europese transportministers in April in Amsterdam afspraken gemaakt.

Het Verdrag van Wenen is het belangrijkste verdrag voor verkeer. Nederland wil samen met andere vooroplopende landen de wetgeving veranderen, zodat de nieuwe ontwikkelingen naar zelfrijdende auto’s ook echt mogelijk worden.

Het Besluit ontheffingverlening exceptionele transporten regelt de ontheffing voor voertuigen die niet in de ‘normale’ categorieën passen. De testaanvrager moet eerst aantonen dat het testen op een veilige manier gebeurt. Daarvoor moet hij een toelatingsaanvraag doen. De RDW kan daarna ontheffing geven.

Vragen over route & navigatie:

Een van de WEpods gaat rijden tussen het station Ede-Wageningen en de Wageningen Universiteit. Vooralsnog zijn dit de enige twee stops op deze route. De andere WEpod rijdt rond op de Campus. Ook deze heeft een aantal vaste stops. Wel zit er een STOP-knop in het voertuig die gebruikt kan worden, bedoeld voor noodgevallen.