Automated driving, what is it and why should we care about it?
Well, answer is simple, its about safety, about comfort and about efficiency.
Seat belts were the first safety systems built in cars and to this date have been proven to be very good in saving lives. Modern Cars are built to absorb energy of an impact and deform in a controlled manner minimizing the impact that we humans experience in a crash. Airbags were later introduced to further minimize impact of our bodies during the crash. This is all part of what’s know as passive safety. But there is only so much these systems can do.
The next step in vehicle safety was active safety which like the name suggests, these systems try to actively warn, avoid or mitigate the effects of accidents by supporting the driver. They are usually called collectively Advanced Driver Assistance systems (ADAS).
How many people would admit to driving and playing with the stereo, only for a few seconds and then suddenly looking up to the horror of a car breaking in front or even coming towards you.
Lets not forget our Calling, SMS, tweeting or Face booking while driving….wouldn’t it be great if you could do that legally and safely while traveling?
The fact is that every study ever made on why we humans have accidents has resulted in a simple answer, because we are in control!
The truth is that we humans, even the best drivers, get tired, are distracted or even make bad decisions. Machines don’t!
They tirelessly execute their mission and can respond much faster than any human can. That’s not to say that machines are perfect, far from it since they are only as good as their programming.
If you think about it, it takes us humans several years from birth before we can walk, talk and safely cross the road unaided. In comparison we can assume that a machine would require a similar amount of time to achieve anything close to our experience and be able to deal with the world.
However, we are trying to achieve a better result in less than half the time!!!
Similarly, Another important aspect is efficiency. As human drivers we collectively waste fuel because of our driving style or even because of our individual sense of danger. For example, we tend to brake too much even when not necessary or accelerate too much. The side effect of this is higher fuel consumption but also observed as “spook files” otherwise known as ghost traffic jams.
Ever wondered why you may be driving on the highway and suddenly you come to a full stop for a few minutes only to return to full speed a few minutes later? Well, most likely someone right at the front has hit the brakes a bit too hard and this effect has traveled (i.e. everyone behind has braked) like a wave leading to you having to come to a full stop.
Automated driving is essentially an advanced fusion and extension of active safety systems aiming at assisting the driver in dangerous situations or as I would term it, correcting our mistakes and even taking over the driving task for example on a long highway drive. Automated driving is also about reducing congestion and optimizing efficiency of vehicles in terms of fuel be it electric or fossil.
In order to achieve this goal there has been a global consensus on a road map defining the levels/steps if you like that lead from manual driving (level-0) to fully automated driving (Level-5) i.e. no steering wheel required.
I think it’s fair to assume that automated driving with no steering wheel is our ultimate goal but something that will realistically take time.
Everybody’s has heard (I presume) of the famous Google car that is capable of driving without a driver!
So is that it, are we done! The answer is no! far from it.
Google among others have presented the world with a series of publicity stunts demonstrating an autonomous vehicle but how can we verify the true reliability of such a system? There are no or limited publications, there are no details about the technology used, how its used, the algorithms and all those important aspects of the vehicle control that would help us prove what it really can and can’t do, how safe is it etc etc.
And what about liability, the law? Who is responsible when a car with no driver is involved in an accident? Can we take a machine to court?
The Geneva convention clearly states that human must at all times be in control of their vehicles or animals….how do we make sure this is the case safely or is it better to let the vehicle be in control??
How do we change the law safely?
What’s the safest way to react to a situation? When should I react?
What about traffic management or the current infrastructure setup which was designed for manual driven cars?
And how does vehicle automation compare to aviation, agriculture, maritime and other domains. Can we learn from these?
You also have to wonder, is this what people want? When do people want automation?
Questions, questions and more questions….
For this purpose the Dutch Automated Vehicle Initiative (davi.connekt.nl) was born spearheaded by TUDelft, TNO, Connekt and RDW with the support of the Dutch ministry of infrastructure and the environment to implement, study and demonstrate automated driving on public roads in order to better understand the technologies and the influence of these on the current infrastructure, other drivers and try to answer the key questions.
In this set of blogs I hope to address the main aspects of vehicle automation and the issues faced by researchers, lawmakers and industry in making these technologies widely available in the broader consumer market.
I will also try to show the diverse groups involved in researching these topics.
If you’re interested in learning more, keep an eye on the blog or even better, get involved by taking up a research project in this field or contributing in some other way!