Electric (self) driving capability

posted in: BTA blog | 0

sup_bannerAutonomous cars on the track

In connection with the recent advances of electric cars in the past few years, it is worth recalling the golden age of the automotive industry. There is a lot to learn from the past. At the turn of the 19th and the 20th centuries, only some ten thousand vehicles were in operation on the roads of America. Most of them used steam-powered, external combustion engines operating on the well-known principle of the steam engine; however, internal combustion gasoline engines were already present. More surprisingly, one third of the cars were electrically powered models. So where did all the electric cars go for a whole century?
To answer this question, we must be aware that all three types of cars at the time used to be very expensive and quite unreliable, too. And compared to horse-drawn carriages, they did not really have significant advantages in terms of speed or driving range. Also, each of the three types had its own specific issues. However, Henry Ford – committed to gasoline engines as he was – managed to construct the first truly usable car in his factory in 1908, the Model T. In the coming years, numerous inventions were introduced to solve the issues of internal combustion engines. Ford was the first in the industry to use a production line for manufacturing, achieving a significant drop in prices. Eventually, gasoline-powered cars became the norm due to their effective mass production. The production of electric cars ceased in 1920, and it was gone for a long while.

A closed industry

The events that took place at the beginning of the 20th century had a decisive effect on the automotive industry, which lasted for a whole century. The internal combustion engine ruled supreme over all other technologies, and severe limitations were placed on those wishing to enter the industry. Developing a new car takes an enormous amount of time and involves great costs. Only a large, coordinated group of well-trained engineers can hope to solve the technical issues that are sure to arise during such a development project. When the new product is ready for manufacturing, there is also the issue of marketing. As a new brand, it is not well-known, has no past and is not yet proven, i.e. even if it is a worthy competitor on the market, no one is aware of this fact. Making the public familiar with a new brand incurs enormous costs spent in marketing. And even so, success is never guaranteed. This is a risk not many are willing to take.
The story of American car manufacturers is a good example. The USA has a huge internal market, with an economy built on entrepreneurship and competition, and a social structure that encourages innovation on all levels. In spite of all this, it was in 1925 when an American automotive company was last founded. Its name was Chrysler. Those wishing to start a car manufacturing business following that date had no real chance at it. The reason: the cards have already been dealt.

Slow development

However, due to the closed nature of the automotive industry, the players became too slow, satisfied with things as they were. This is inevitable, if there are no new competitors. Of course, there is some competition going on between existing manufacturers, and cars did in fact undergo a significant development in the course of a century. However, all manufacturers were in silent agreement about one thing: let’s leave the basic propulsion technology alone. Modernizing was OK, but developing a radically new technology would have taken an enormous amount of time and money. Also, they were not going to discard their gasoline- and diesel-powered engines just like that. They could’ve developed a well-functioning electric car, since they had all the means required: money, research base, engineers. But they decided not to, as they were used to the fact that the automotive industry didn’t have to face challenges from the outside, and they could easily preserve the status quo. They procrastinated until a 21st-century Henry Ford came and conquered. His name is Elon Musk, CEO of Tesla.

A number of issues

To develop a viable electric car, Musk and his team had to overcome significant difficulties. A few years ago, battery-powered vehicles were infamously weak. Today, Tesla’s most powerful model can accelerate to 100 in a matter of 3 seconds, which is comparable to the performance of racing cars, but even less dynamic models can cope with this task within 5-6 seconds. They have a range of approximately 400 km with a single charge (although this value heavily depends on the speed), which means that their range is also comparable to gasoline-powered cars, even if lagging behind them a little bit. If a Tesla is used solely for the purpose of everyday travel needs in a city, there is no need for refueling: all we need to do is connect the car to the charger. However, for longer distances between cities, a charging network needs to be in place. Tesla took enormous steps to comply with this requirement: their Supercharger network is developing fast. Using only these charging points, almost the entire USA and Western-Europe is accessible today, and the network is growing quickly.

Electric car for the masses

The business model of Tesla can be summarized like this: build expensive cars first to prove the viability of the technology and to produce an income that enables the company to enter the second phase, i.e. manufacturing cars for the wide public. Their first car developed fully in house was Model S, released in 2012. The next type, Model X was introduced in September. These are quite expensive cars which can cost 70-140 thousand dollars depending on optional accessories.
They have the expensive cars now, the proof of concept has been widely accepted, therefore it is time for Tesla to start concentrating on its primary mission: producing an electric car for the wide public. To achieve this, they have to reduce the cost of batteries drastically. That is why they have already started to build their own battery production facility, the Tesla Gigafactory. The facility will truly be gigantic: more lithium-ion batteries will be produced here than the amount produced in 2013 worldwide. With such a huge economy of scale, a 30% drop in production costs can be achieved. This way, Model 3 can be released as early as 2017, with a planned price tag of 35 thousand dollars. This is still not cheap, but in developed countries it will surely be accessible to a much larger customer base.

Better in every respect

If this plan succeeds, there will be no more arguments to support gasoline-based vehicles. In terms of performance and range, they are already in the same league. The engines of traditional cars have hundreds of parts, while electric engines contain an order of magnitude less than that, which leads to the elimination of a number of possible issues, making the latter much more reliable. Electric engines are also clean, which means that there is no need for replacing the oil, for example. Their operating costs are much lower, even after the drop in oil prices. The engines of traditional cars are the product of a mature industry, which means that there is little room for improvement, while batteries still have a huge development potential. Electric engines are much more efficient, less energy is wasted, and there are no harmful emissions.
We take it for granted that cars are smelly and noisy. We are so used to the fact, we forget that this is not at all inevitable. We could drive silent cars with no emissions, without producing smog in cities. However, the paradigm shift is approaching fast. When I see a car with a black trail of smoke behind it, I always think: this is the legacy of the past, the memento of the last century in the present. More and more people will be convinced that internal combustion engines are obsolete, a thing of the past. Something we can soon get rid of.

Missed development opportunities

As mentioned earlier, the entry barriers of the automotive industry have made car manufacturers slow and unresponsive. Missed opportunities in technology and potential innovations accumulated, but were never utilized and did not appear in vehicles. Manufacturers could afford this due to the lack of real competition.
Tesla could break the barrier, because the amount of missed opportunities in development have reached a critical mass. This made it possible to create something new and cool which gets the attention of the public. And this is not just about the electric engine itself.
If you try a Tesla, you will surely understand what we mean by the lack of competition between traditional car manufacturers. It is undeniable that cars have evolved significantly. But this is just a portion of what would be theoretically possible in the presence of real competition. Even avid fans of gasoline-powered cars are usually surprised how good a Tesla is once they tried it. There are hundreds of testimonials on the Internet, full of praise of the new car.
The biggest failure of traditional manufacturers is not that they failed to develop the electric car in time. A missed opportunity it is, but not the most important one. I am convinced that it is in the software where most opportunities have been missed. This is not surprising, as these companies are not based in IT: they have a mechanical background.

It is the software

By developing the electric car, Tesla challenges traditional manufacturers in a field where their experience with internal combustion engines does not mean an insurmountable advantage. They are forced to compete on a level field. In this environment, Elon Musk (creator of PayPal, among others) can demonstrate how underdeveloped the software solutions of traditional cars are compared to what’s theoretically possible. Using the hype around its electric car, Tesla can enter the automotive market to utilize the real market niche that is the software.

Mobile phones and cars

Let us recall the story of the BlackBerry mobile phones to draw an analogy here. In the middle of the first decade of the 21st century, the direction of development for smartphones was still uncertain. BlackBerry produced phones with mechanical buttons. In fact, a full numerical and alphabetical keyboard was integrated into the device. When touchscreens appeared, this approach proved to be a dead end. Why? Because we don’t need the keyboard most of the time. They take up valuable space, reducing the potential size of the screen.
When we use our smartphone today, it feels natural that only those buttons appear that are needed at the time. Letters, numbers, or whatever is necessary. When we receive a call, the Accept and Reject buttons. It feels natural that completely new buttons appear onscreen after downloading a new app, enabling us to control the new functions. We take it for granted that software applications are regularly updated with new functionality.
However, when we get into our cars, all of this goes away. The car has exactly the same functionality as when we bought it years ago. We do not expect new buttons to grow out of the dashboard, and don’t care that some quite useless buttons take up space. This gives a precise picture of the obsolete approach used by traditional car manufacturers.
In contrast, Teslas hardly have any mechanical buttons or switches. Only frequently used controls like the turn signal lever, or emergency controls like the horn button or the hazard flasher switch have been kept. All the other functions are implemented on the huge 17-inch touchscreen. This enables the driver to customize and access any function from basic ones (such as the headlights, ventilation controls or the sound system) to more complex controls. This screen also serves as the interface for any future functions to be released later. Regular functionality updates similar to the ones on smartphones can finally become a reality in cars as well. But will this potential be realized?

Expanding functionality

Updating the software of a traditional car is a complex and painful process. It usually costs money and can only be performed by a trained technician. Often even technicians refuse to perform the update, or they do so only at the risk of the owner, as software updates sometimes lead to component failure. Also, the updated software usually does nothing more than optimize the operation of the engine (or the detection of a lab environment for emission checking), and there are no truly new functions.
With Tesla, software updates are a different thing entirely. The company releases a new software version every 1-2 weeks, automatically downloaded by vehicles. New functions appear in the software, expanding the capabilities of Tesla cars week by week. It’s as if new buttons were growing out of the dashboard. On the one hand, this is surprising. On the other, this should be the norm today, just like in the case of smartphones. Traditional car manufacturers clearly have a debt in terms of software. Let’s see some of the new functions Tesla offered recently.

Autonomous cars

Version 7 of Tesla’s operating system was released in October. It is one of the more exciting releases, as it incorporates the Autopilot package. New functions include Autopark, Autosteer and Auto Lane Change. Other brands also offer high-end cars with similar capabilities, but Tesla have gone further than anyone else in making the car truly autonomous.
A good example of this effort is the Autosteer function. Other cars with a similar function usually do not interfere with driving aggressively. They warn the driver if the car is at the edge of a lane or if the following distance becomes dangerous, but the control basically remains in the hands of the driver. Tesla’s Autosteer function takes a step further. When activated, the car essentially drives itself. Typically, this can be uses on motorways. Using data from the GPS and integrated sensors (radar, ultrasound, camera), it keeps the car in the middle of the lane, while maintaining a safe following distance. It detects obstacles and free spaces, braking and accelerating automatically. The function is also useful in traffic jams. Staying in the same lane, the car automatically starts when possible, and stops if necessary. If the software cannot decide on its own, a sound alarm warns the driver to take control.
This is very interesting even though Tesla emphasizes that the Autopilot package is a public beta test version to be used at the driver’s own risk. Drivers need to remain engaged and aware, as there are situations when they need to interfere, although not frequently. However, cars memorize these situations, and they send all related data to the company the first chance they get. The company analyses the information and uses it to improve the Autopilot package. So when a Tesla encounters a situation it cannot handle, the data from this situation will help improve the entire fleet when they download the next update. Eventually, as their experience in traffic grows, the cars will become more adept at autonomous operation. And let us not forget that just a few months ago these cars were not at all capable of automatic driving.

The brain of the car is the software

It is clear now why the software is such an important part of a vehicle. A lot of people believe that one day the self-driving car will be created by one of the companies, and then it will be available to the public as an off-the-shelf, out-of-the-box product. This reflects the obsolete approach of traditional manufacturers. In reality, things will be quite different. These future cars will be out on the market well before they are capable of truly autonomous operation. In fact, the self-driving cars of the future might already be on the roads today, with their full potential still waiting to be realized. The functions to enable self-driving will be incorporated later into their software, with a future update.
Apple CEO Tim Cook said at a conference in October that the automotive industry is facing massive changes. The software will become more and more important in cars, and self-driving capability will be a key factor. He did not comment on the gossips that Apple is developing its own self-driving electric cars. He only said that their goal is to offer the equivalent of the iPhone experience in cars. However, the direction they are taking seems quite obvious as more and more engineers and researchers in self-driving technology are employed by the company. At the same time, Google has been experimenting with self-driving cars for years now, with 2 million kilometers of self-driving experience behind them.
It is obvious that large technology companies are becoming more and more interested in producing the car of the future. Tesla has some advantage with its electric cars, but given the heavyweight opponents it has to face in the future, its long-term success is still uncertain. Tim Cook was certainly right in stating that the automotive industry is facing massive changes. The last time this happened was a century ago. As we can see, the card that were dealt at the beginning of the 20th century are beginning to lose their meaning. This round seems to be over. New players are eager to enter the game, preparing for the next century. The new cards are being dealt right now. An exciting era is coming.