Self Driving cars, or autonomous cars as they are referred to, are vehicles that are capable of sensing their environment and navigating through it without any human input. There are a number of technologies that go into building a self driving car including Radar, Lidar, GPS, Odometry, and computer vision. A distinction should be made between autonomous cars and automated cars. Autonomy means self governance and the power to ply without human intervention. Automated cars have been around for some time and are typically built to depend heavily on artificial hints in the environment (e.g., the cars that run on magnetic strips in the Jurassic Park movie).

Self Driving cars can be classified into 6 Levels based on classification provided by the Society for Automotive Engineers International (SAE International)

● Level 0: Automated system has no vehicle control, but may issue warnings.

● Level 1: Driver must be ready to take control at any time. Automated system may include features such as Adaptive Cruise Control (ACC), Parking Assistance with automated steering, and Lane Keeping Assistance (LKA) Type II in any combination.

● Level 2: The driver is obliged to detect objects and events and respond if the automated system fails to respond properly. The automated system executes accelerating, braking, and steering. The automated system can deactivate immediately upon takeover by the driver.

● Level 3: Within known, limited environments (such as freeways), the driver can safely turn their attention away from driving tasks.

● Level 4: The automated system can control the vehicle in all but a few environments such as severe weather. The driver must enable the automated system only when it is safe to do so. When enabled, driver attention is not required.

● Level 5: Other than setting the destination and starting the system, no human intervention is required. The automatic system can drive to any location where it is legal to drive.

Looking at the history, the initial seeds were sown as early as 1920s with promising trials happening in the 1950s. Only in the 1980s did teams of engineers from Carnegie Mellon University’s NAVLAB and Mercedes Benz’s EUREKA project crack the code for the world’s first self sufficient and truly autonomous car. The real breakthrough however, came in 2011 when the state of Nevada passed a legislation allowing self driving cars to be plied on its roads after intense lobbying by Google, which had already conceptualised and tested its SDCs under the Google X project.

A few of the players in this space are:

There are a lot of players in this market. Some of the big names in this space are not just automobile firms but also technology firms such as Apple, Google, Baidu, and Intel. The sensors, smart engines, radar technologies, entertainment systems, and a whole ecosystem of apps are few of the entry points for the names mentioned above. There is a huge upside for some of the existing players as they have a strong foothold in the industry in general and the pure technology firms are either partnering with the automobile firms or are acquiring them. The overall market size of the self driving car industry is expected to reach $42 Billion by 2025 and soon after overtake the traditional automobile industry in terms of market size and growth.

Self driving cars have given way to the idea of replacing human driven vehicles with autonomous ones. A few of the other autonomous vehicles being tested right now are trucks (Otto: and busses which are expected to hit the roads next year.

The infographic below is a sneak peak into the workings of a self driving car:

Self driving cars are prone to accidents and fatalities like any other vehicle and are under high levels of scrutiny given the controversial nature of such accidents. A report from The Next Web states that more than 94% of all vehicle crashes are man made and moving completely to a world of self driving will bring this number down drastically. A quick look into the crashes involving self driving cars reveals that Google’s self driving cars have been involved in 5 reported accidents whereas Tesla’s Model S has been involved in a fatal crash. Up until that point in time Tesla had a clean record of 0 crashes for 130 Million miles way higher than the national average as reported by the NHTSA. The important question to be answered is how do we make self driving cars smart enough to understand the implications of crashing into a living being, another vehicle, or even the guard rail on the side of the road to save the driver’s life. What is

even more complicated is to save more than one life on the road at a risk of bringing harm to the driver and the passengers.

Moral Dilemmas

Imagine a self driving car carrying one passenger through a two-lane street when three kids suddenly run out in front of the car to retrieve a ball from a game they were playing. The brakes are not working and the other lane is under construction in that specific segment of the road, so swerving to the side would most probably kill the passenger of the car. However, if the car does nothing, three kids will most likely be killed. What should the car do?

This situation, which is getting some attention of the media in recent years, derives from the ‘trolley’ problem, philosophical thought experiment: imagine you are watching a train go straight into five people who are tied to the tracks. You do not have time to do anything other than activating a switch to make the train switch tracks, but there is a person on this other rail also tied to the tracks. The moral dilemma is that, even though you are saving five lives at the cost of one, you would be actively killing that one person by activating the switch.

While the scenario described with the self driving car may seem extreme, it is an important consideration for car manufacturers. What should the car be programmed to do in that situation (if, at all, it should have this in the code)? Always prioritise the passengers? Choose an outcome with less casualties? Prioritise whoever is obeying

the law (e.g. pedestrians crossing on a green light)? Should this choice consider the age of who is involved — prioritising children, for example?

To gather people’s opinion on this topic, MIT developed a website called Moral Machine. In this website, users are presented with 13 scenarios in which they are required to make a choice among two outcomes. These scenarios range from choosing the passenger over pedestrians, selecting one group of pedestrians to die over another one and different characteristics of possible victims: fitness and age, for example. It also provides information about who is following the law: some pedestrians may be crossing on a red light, for instance. An example of a scenario is the following:

On the first choice, four women pedestrians are crossing a red light and will be run over. Two of them are executives and two of them are fat. On the second choice, the car will

deviate from the women and hit a concrete barrier, killing all four male passengers: also two executives and two fat ones. The user is required to select the best choice.

After you go through all the 13 scenarios, there is information about the rationale of the user’s choice versus what other people in general answered:

An interesting consequence of this dilemma is that self driving cars would never be popular if people knew that the car would not protect the driver in case of an accident. Moreover, the answers people provide should be taken with a grain of salt: people answer these questions with the lens of an external observer, and not as a passenger of the car.

Future of the Industry

Many social, economic, technological and governmental trends will together shape the future of mobility. The way people will move around both urban and rural environments is set to drastically change in the near future. Existing companies, such as Uber and Didi Chuxing are already changing the traditional methods with new technology, connectivity and autonomy.

Cities and countries will embrace these changes in different ways that cannot be predicted from a local point of view. However, the hub cities around the world will most likely set the standards of the advanced mobility model. It will first and foremost start with the people because the individual traveler will dictate the speed of change of this new industry.

A McKinsey Mobility Report has forecasted the usage of autonomous vehicles as part of a two-scenario model: a high-disruption scenario and a low-disruption scenario. The high-disruption scenario is based on fast, comprehensive and enthusiastic adoption in regards to regulatory challenges, safe and reliable technical solutions, consumer acceptance and willingness to pay. Whereas the low-disruption model assumes gradual, incomplete and limited adoption in regards to the challenges mentioned before.

In the chart above, we can see that the first signs of self-driving cars usage and adoption will not happen until after year 2025, even in the high-disruption scenario. Year 2035 will either mark a scenario where over 80% of cars are fully autonomous based on availability in popular consumer models or a scenario where technical and regulatory barriers will delay popular adoption. Moral hazards will likely be overcome once the technology is semi-perfected, because the consumer will quickly realize the speed at which the technology will improve. It will be such that, if a self-driving car is involved in a car crash, the algorithm automatically learns from its mistake and updates all other self- driving cars in use. This is in comparison to a human making a car accident and learning for himself / herself, the autonomous cars will be able to learn seamlessly from each other.

At a government-level, politicians and reformists will want to anticipate theses new trends and craft regulations that are consistent with consumer-friendly developments that respect a minimum ethical and moral standard. Strong partnerships between the public and private sector will likely product the most innovative and best solutions and will be viewed as the backbone of a successful transition. Private companies who will have collected massive datasets and have developed the technology will need to be regulated in such a way that they protect the consumer. Governments, as the middleman, will provide the correct structure to maintain citizens safety and society’s overall health.

Solving this problem right will result in a significant competitive edge for cities and countries around the world. This shift will result in reduced pollution, traffic, deaths and increased productivity. It is an opportunity to impact and improve quality of life for billions of people across the world.

As shown below, it appears that most people are optimistic about self-driving cars, and as the technology keeps improving, the trend will only evolve. How do you feel about it?


Paper by: Paulo Manso (, Chris Abi-aad (, Abhiram Muddu (


MIT Technology Review

The Guardian

Scientific American

World Economic Forum


The New Yorker Moral Machines


The Next Web

Market Watch