The Thinking Car

We were at Pier 27 in San Francisco, at the city’s edge that looks over into the blue waters of the Pacific Ocean that you can see clearly right up till about 9 at night, which is when the sun sets there in this season. This was early morning and on the driveway were numerous cars that could be taken for a ride. They were some of the latest models of Ford with a range of features that assist drivers. But one of the things that technology cannot give is courage. I had never driven a car with the steering on the left and didn’t plan to start now. The previous day at Palo Alto, or Silicon Valley as we know it, I had seen a signboard which said that the fine for breaking a red light was $336. In Indian rupees, that would be a little over Rs 20,000. Traffic offences there can be crushingly expensive, and as a driver unfamiliar with both the land and vehicle, I decided not to test my luck. Also, just some time earlier, I had sat with another driver for a spin and we had got lost because while we had been provided a map with clear directions, at a junction it asked us turn left to Hayes Street but a board there said ‘no left before 9 am’ and the time was 8.55 am on our watches. We couldn’t take the turn, drove through half of downtown, beautiful as it is with streets that are like rolling hills, and then just about managed to get back without a nervous breakdown. I definitely didn’t want to drive.

That was how at Pier 39, instead of being in a Mustang, I found myself a little further away at the parking lot, where a demonstration was on. I saw a few square lines and a short barrier at its end that indicated an imaginary garage. A man got out of a car before the lines, held a keychain that was a remote control, pressed a button, and the car moved by itself into the space. Just to be clear, he was not manoevring the car with the remote. The button only gave it a command to park and the car did the rest. To most people, parking is a distasteful necessity in the driving experience, but technology is taking it out of human hands. The system uses multiple embedded sensors that measure the distance around the car to find its position, and then the car takes over the steering to fit itself into a slot. When the driver wants to take it out, if the car is squeezed between two other vehicles, he can just press a button on the remote again and the car will come out by itself to where he is. All this is something like the first stage of a car becoming a thinking robot that carries people in it. We are now entering an age when cars will drive themselves. In the jargon of engineers, the term is ‘fully automated’.

A futuristic is a person whose job is simply to look at what the future will be. That might sound like a hobby, but multinationals, especially those driven by technology, have people on their rolls to do that. Intel, for instance, has Genevieve Bell, even though she herself professes to be uncomfortable with the label. She was one of the speakers at a panel discussion on automation at Further With Ford 2015, an event organised by the motor company at Palo Alto to discuss trends that are shaping the world.

Bell is an anthropologist who says one of her jobs is to think about what human beings care about—their passions and frustrations—and to help design technology around those things. “I spend most of my time hanging around with human beings and getting a sense of what makes them tick,” she says. A fully automated car might become a reality soon, but in the imagination of mankind it has been there for some time and one image she put up at her presentation was an advertisement in Time magazine that spoke about being in one on a highway in California. The advertisement’s year: 1958.

Man’s efforts at automation go way before that, say about 300 years. One of the earliest instances was in 1731, when a Frenchman named Jacques Vaucanson built a mechanical duck. “He was the first person to use vulcanised rubber,” in Bell’s description, “He made a digestive system in this duck. With water, so it gurgled. It would be taken to the courts of Europe, you put food in its mouth, it digested it. And then it did the other thing that ducks do.” So it had a container inside full of duck faeces collected before every event and the demonstration included the duck’s excretion of it. “He used all the technologies of his day to make something that looked real. It was the first time that human beings could use technology to make something that could effectively come to life,” says Bell.

As full automation in cars become a reality, it will throw up complex questions like how much autonomy a machine should have, who will regulate it, what it should look like, and what its form and function ought to be. And whether, if you give a machine autonomy, will it then become sentient in some sense and have an inner life of its own? Bell has an interesting anecdote of how these questions can change with cultures. She was once at Intel’s facility in Japan and saw a sign that said ‘Robot Zone. Autonomous Robots Two Metres in From the Kerb’. The sign was intended for robots! And Bell recounted an amusing exchange with her Japanese colleagues. She asked who the robots belonged to, and was told that they were autonomous, seeming to suggest that they existed as independent beings. “[I asked] how did they get here? ‘Autonomously.’ What are they doing? ‘Being autonomous.’ There was something here about the notion of what [it means] to grant autonomy.”

We already live with machines that are autonomous to some level. Bell’s example is of her ATM machine, which wishes her on her birthday. Or robotic vacuum cleaners which can be bought online even in India. According to her, when self- driving cars become a reality then there will be debates and discussions on their use among different classes of consumers, governments, cultures and even members of the same family.

Ken Washington, Vice-President, research and advanced engineering, Ford Motor Company, who spoke before Bell, had explained just what it takes to make a car that drives itself. The first stage of making a car intelligent is for it to assist the driver and that goes as far back as the 1920s with automatic transmission or the 1960s when cruise control was introduced. These technologies went under the nomenclature of Driver Assist, which informs the driver of something that is happening in or around the car to help them drive better. An example would be collision warning. The next stage was for cars to become semi-autonomous, and in this, the vehicle actually takes some control, operating some aspects of the throttle, braking mechanism and steering on its own. So when it apprehends a collision, the car does not just warn you but also applies the brakes to stop in time. These technologies are present even now in many cars. The remote parking that I had witnessed was one. Fully autonomous cars, the next stage, are an entirely different ball game, calling for huge leaps in development.

To drive all by itself, the vehicle needs two prerequisites. It must have the ability to scan its environment in 3D. It also needs to have a pre-existing 3D map of the environment under normal static conditions. It will then compare what it is seeing with the pre-existing map, and use that to move. “This is not hype, this is going to be a reality. Within the next five years, maybe even faster. Where are we today? You can buy a car today that has semi-autonomous and driver assist technologies in it and those capabilities are going to increase and become more and more capable in future,” says Washington. Ford already has a fleet of research vehicles on its Fusion platform that are testing full automation. These cars have four horn- like appendages on top called LiDARs, which do the 3D mapping. Washington says a breakthrough has been made. “It just transitioned from a research programme to an advanced engineering programme. That is important because the research programme is taken to answer the fundamental question—is it possible? Does physics allow this to happen? The answer is a resounding ‘yes’. The next question in the advanced engineering phase is: how do you do it? What are the combinations of hardware and software needed to bring this to a product? We are in that phase today,” he says. The phase after that will be product development.

Cars that drive themselves will pose numerous questions. Human drivers can, for example, read the intent of others very well through eye contact and body movement. If a pedestrian is waving, humans know whether it is to greet a friend or for help in an emergency. Such issues need to be interwoven with technology while developing a self driving car. Or what happens in a society where people go from being drivers to just riding in the back seat? What will they do with their time? How will the nature of services they use during the ride change?

The answers will come once the conversations start around these issues. Autonomous vehicles are one element in how auto companies are redefining the way they approach the mobility of human beings. A number of factors are responsible for it. At the Further With Ford 2105 event, a whole slew of initiatives and technologies showed just how much reshaping was being done.

Ford’s CEO Mark Fields spoke about how the company is now focusing on flexible use and ownership, and multi- modal urban mobility solutions. In plainspeak, that means things like using technology so that, akin to car-pooling, there can be car-sharing. “We are providing customers [with the facility] to rent their vehicles to others for short-term use, and that way they can save and earn some money,” he says. The company is starting with 14,000 customers in six US cities and 12,000 in London. Other initiatives include making electric bicycles that people can use along with cars while commuting. Or regular vehicles equipped with sensors which, as they move, collect information like parking availability in cities and pass it on so that everyone benefits and can plan by it.

The futurist of Ford, Sheryl Connelly— her actual designation is ‘manager, global consumer trends and futuring, Ford Motor Company’—says that one of her jobs is to read what is shaping the world so that the company can plan for it. “My role is to ensure that that conversation stays broad enough so that we don’t miss anything,” she says. She would therefore analyse trends that range from the impact of increasing population to the effect of human beings living much older now.

She says there will be an increasing number of mega cities with populations greater than 10 million and most of them will be in Asia. “It is very difficult to move about in a mega city. The average daily commute in Beijing is five hours a day. In the US, there have been traffic jams in the last two years that lasted 24 hours. The real problem is that nobody tells the cars that are 10 or 20 km behind them and they wait in vain. If the world population does grow to 9 billion, it is possible that there could be 4 billion vehicles down the road. Who will that serve if none of those vehicles move?”

Hence this change in focus to mobility and designing technologies to address that. Take the fact that as people age and live longer, they find themselves unable to drive. This is in turn prompts auto companies to put in more features to help them remain mobile and drive safely. Ford, for example, has just launched advanced cameras in cars that offer 180º views both at the front and back of the vehicle, warning drivers of any obstacle around. “These features are what are going to get us closer to autonomous cars,” says Connelly.

About The Author

Madhavankutty Pillai

Madhavankutty Pillai has no specialisations whatsoever. He is among the last of the generalists. And also Open chief of bureau, Mumbai  

•