Remember how we asked our parents: "You really didn't wear a seatbelt in the car when you were kids?" – In less than two decades, your grandchildren will be asking: "You really drove your cars by yourself?" Their eyes will be wide, as they wisely tell us how dangerous and crazy that experience must have been.
Through the lens of the future, today's world will be hard to explain, but this is nothing new. Progress has always been like that. However, the most amazing thing is that the time period between these major changes is shortening at an unbelievable pace. A few years ago, you had to explain to your neighbors, why you bought an electrical car. Today, in Silicon Valley, you have to explain why you didn't.
1. Self-driving cars will change our lives
Research on self-driving cars began right after the Millennium. There were several reasons to start such research:
- In cities, 20 to 30 percent of usable space is taken up by parking space. 97 percent of the cars in the world are parked somewhere right now – only three percent are moving! However, with self-driving cars, we would need fewer cars and parking spaces.
- Every year, roughly 1.2 million people die in car accidents.
- Every person spends thousands of hours behind the wheel, time that could be used to do something more productive.
2. Vehicles make intelligent decisions
The first challenge for self-driving cars was carried out in 2004 in the Mojave Desert. The idea was for teams to overcome a 150 mile long course with their self-driving cars. None of them succeeded – seven miles was the record. One year later, five out of the eleven teams succeeded, with the Stanford University team as the winner.
Stanford's team did not focus too much on the hardware (the car), but rather on the software (the brain). Instead of trying to teach the car every possible situation, they let the car learn through behavior cloning – simulating human reactions in various situations. This was particularly important for the following "Urban Challenge" two years later, where the route led through urban cities. The vehicles had to make intelligent decisions in real time based on the actions of other vehicles; Interactions, such as maintaining precedence at a 4-way intersection.
Sebastian Thrun, who led the Stanford team, was asked in 2009 by Larry Page from Google to build an autonomous vehicle for the tech giant. Within 18 months, the first Google car was developed. By now, Google has presented the third generation: the first in-house designed, driverless car without a steering wheel or pedals – the so-called Google Love Bug.
3. Superior learning through artificial intelligence
Artificial Intelligence (AI) is currently a very hot topic in the Silicon Valley. Software for self-driving cars relies on AI through deep learning algorithms – software that writes software by simulating the work of a human brain and also by learning from their mistakes. In other words: by using adaptable Algorithms instead of code.
Self-driving cars adapt very quickly. While humans tend to learn from their mistakes over an entire life span, driver-less cars share their experience with all other cars. Therefore, their capacity to learn multiplies itself constantly. Every month, the 'intelligence' of self-driving cars gets ten times more advanced! According to Sebastian Thrun, the fleet of self-driving vehicles will reach 'absoluticity' in one year – meaning that no more mistakes will be made.
4. Smart cars will be affordable for all
According to the 'Wall Street Journal', Volvo, Tesla, Nissan and Ford expect to have autonomous vehicles within the next five years. Therefore, the pressure is rising in the car industry. Machine vision is crucial for autonomous vehicles and Nvidia produces the graphical processors required for it. This is a company that has its origins in the gaming industry, and is practically synonymous with fast computers. It is no surprise that the computing ability in a self-driving car is comparable to one hundred and fifty iMac Pros. Danny Shapiro from Nvidia said that the production cost for the computer needed is about 200 Dollar. In August, Delphi and Mobileye announced the development of a fully autonomous system for cars in 2019. They want to produce an off-the-shelf system device for all types of cars, even for small trucks.
5. More reliable than humans
How long will it take until driving a car yourself will be punished like drunk driving? Perhaps you will have to explain yourself in front of a court because you caused an accident. Eventually, you won't be able to find insurance that covers driving yourself, as machines become more reliable than humans. Like it or not, this is the direction the world is heading.
This being said, turning control over to software could lead to new hacking vulnerabilities and other hazards - liability issues that companies can't ignore. Although NuTonomy launched the world's first self-driving taxi service in August and Uber started its pilot with self-driving cars in Pittsburgh earlier this month, security issues have to be solved before declaring self-driving cars as a new standard.
5. New concepts for cities
Cities around the world will raise taxes on non-electrical cars and switch to self-driving cars to free up parking spaces in densely populated areas. Those self-driving cars, though, won't be owned by your. You will have to hail them - access becomes more important than ownership.
Otherwise business people would ride in their Tesla to the city and let the car circling around on its own during their meetings …
There will be self-driving buses as well. Who would have ever imagined the Swiss Post operating an autonomous bus? Aside from the fact that the idea was born during a Silicon Valley trip, the autonomous bus in Sion is already smaller than a regular bus – perhaps in the near future, they'll merge it with Google's Love Bug and we'll be able to ride in cabins for four.
New developments will only be used mainstream when they add value to our lives. With key factors like security, urban space and time-efficiency self-driving cars will prevail. Soon, self-driving will be synonym to driving a car by yourself and not the other way around. Within ten years, our roads will be full of autonomous vehicles. In Mountain View, there are probably already more self-driving cars on the road than Tesla cars. The new normal is already here! (Damir Bogdan, Mountain View)
Über den Autoren
Damir Bogdan, founder of Actvide AG, consults companies on digitalization and innovation. He is active in Switzerland as well as in Silicon Valley, mentoring FinTech start-ups. Through his consultancy work he builds bridges between European and US companies. Plug and Play - a major accelerator in Silicon Valley – have mandated him as an Ambassador for Europe as well as an Executive in Residence in Sunnyvale, CA.
Prior to this, Damir Bogdan was a long time CIO & Head of Operations of Raiffeisen, Switzerland's third biggest banking group. As a member of the Executive Board he was responsible for the modernisation of the IT and the implementation of a standard core-banking platform. During his time in the board, the bank started to diversify by implementing a private bank and offering new services to its customers. In his role as Chairman of the CIO's of the European cooperative banks he led several pan-European initiatives.
In his earlier years he had several project und leadership positions within the St. Galler Kantonalbank, AGI IT Services AG und Swisscom. As a member of the strategy team he was responsible for the implementation of an offshoring strategy for Swisscom IT Services AG as well as for the corporate SAP-strategy.
Damir Bogdan is a member of the Advisory Board for Information Management for four universities and a member of several IT committees.
Damir Bogdan holds an Executive MBA of the State University of New York, a Swiss Federal Diploma for Information Management and a Certificate Essentials of Leadership of the London Business School.
Über die Kolumne:
Damir Bogdans Kolumne "Disruptive: a Letter about Innovation" erscheint regelmässig auf inside-it.ch. Wir publizieren die Kolumne in englisch.