The development of connected and autonomous vehicles and their ethical dilemma

The automotive industry is undergoing a major change with the advent of electrification, autonomous driving and connectivity. These three factors oblige automotive manufacturers and suppliers to rethink the functioning of the vehicles of the future and their use. The arrival of autonomous and connected electric cars will change our habits and disrupt the transport sector.

Every year, the growth of the world's population, which will live in urban areas, is increasing, leading to an increase in traffic jam, pollution and energy consumption. Pollution and climate change are partially caused by current vehicles emitting nitrogen dioxide, carbon dioxide and fine particulate matter with diesel engines. Faced with these problems, electric vehicles are quickly imposed as a good solution to reduce CO2 emissions to improve air quality.

During my visit to the World Motor Show, I could see this trend of electric vehicles. Indeed, most car manufacturers have presented electric vehicles at this show and have given a taste of the ubiquity of the electric car in the future.

Electric vehicles have certain advantages:

1. Economical: the engine wears very little, and therefore has a long service life, low maintenance of the electric vehicle because few moving parts.
2. Practical: easier to use, no clutch, no gearbox.
3. Clean: silent, does not pollute.

Today, some vehicles are semi-autonomous with Advanced Driver Assistance Systems (ADAS) which simplify driving and improve the driver's attention. The main objectives of these driver assistance systems are to alleviate the tasks of the driver and assist the driver in his perception of the environment (detection of collisions, overspeed, pedestrians, road signs, etc.). They can thus widen the driver's field of vision, alert the driver to overspeed or crossing the line. These systems can even offer electronic co-operation with anti-collision systems, for example to park the car automatically by measuring the available space thanks to its sensors. Therefore, I think that the arrival of autonomous cars is only a matter of time and will see the light in the near future.

An autonomous car (driverless car or self-driving car) is a vehicle capable of driving on the road without the intervention of a driver. It is a vehicle that is equipped with many sensors like ultrasonic sensors, cameras and lidars in order to perceive its environment. The data collected by the sensors are processed by software and processors to make the data fusion in order to observe its environment and to detect other vehicles, obstacles, signs and pavement limits. Then artificial intelligence algorithms will decide the action to be performed on the vehicle controls (steering wheel, accelerator, brake, etc.).

Statistics show that 90% of traffic accidents are due to errors committed by the driver. This emphasizes the idea of ​generalizing autonomous vehicles in order to reduce these accidents.

Many car manufacturers and web giants, software and electronics are therefore engaged in the technologies of autonomous driving and to develop autonomous vehicles.

Among the companies in the web, software and electronics, we can talk about Apple, Baidu, Google, Microsoft and Nvidia.

• Apple is launching into the autonomous and connected cars sector with the Apple Car. The project is called "Titan", Apple invests a lot in the technologies of machine learning and automation of cars, and would like to present the Apple Car before 2020.
• Baidu, the Chinese search engine allied with BMW to release a prototype that was tested in December 2015 with an autonomous BMW 3 Series that rolled 30 kilometers around Beijing. Baidu intends to commercialize and mass-produce autonomous vehicles in 5 years.
• Google is continuing to improve their Google Car project, which is divided into two types of autonomous vehicles, which are the standard vehicles modified by Google (Toyota Prius, Lexus, Chrysler) and autonomous electric vehicles entirely designed by Google.
• Microsoft, also wants to contribute in autonomous driving, and is working with Volvo to adapt their augmented reality headset, HoloLens, for the cars.
• Nvidia developed an on-board computer dedicated to autonomous vehicles, unveiling the Nvidia Drive PX2 at CES 2016. The computing power of this on-board computer is 8 TeraFlops, which allows to launch machine learning algorithms and to make data fusion of the various sensors embedded in autonomous vehicles.
• Uber has recruited engineers from the University of Canergie-Mellon to set up their autonomous vehicles and will be testing in Pittsburgh.

Automotive manufacturers and suppliers such as Audi, BMW, Bosch, Valeo, Ford, Honda, Hyundai, Mercedes-Benz, Renault-Nissan, PSA, Tesla, Toyota and Volvo are preparing to enter the autonomous car sector.

• BMW unveiled its iNEXT prototype at CES 2016 in Las Vegas.
• Audi had developed several prototypes of autonomous vehicles and plans to market its Auto Piloted Driving system already integrated in its high-end sedans A6, A7 and A8.
• Bosch had worked with Tesla on two Tesla Model S vehicles to make them completely autonomous.
• Valeo announced that their "Cruise4U" system will be embedded in a car and will be tested in the United States, in different cities like San Francisco and Boston, Los Angeles, San Diego, Las Vegas, San Antonio, Miami, New York, Boston, Detroit, Chicago and Seattle.
• Ford keeps on investing in intelligent mobility through connected vehicles and autonomous vehicles, resulting in the creation of a Ford Smart Mobility subsidiary. Ford tested autonomous vehicles in difficult conditions such as during the night and under snow.
• Honda tests their autonomous vehicles in California and also offers semi-autonomous features (trajectory maintenance, automatic braking and automatic speed) on its Civic models.
• Hyundai continues to work on autonomous cars but thinks the arrival of autonomous cars will be delayed.
• Mercedes Benz wants to be at the forefront in autonomous cars, and plans to offer many semi-autonomous features in its luxury sedans. Mercedes Benz will also embark on the luxury limousine service and compete with Uber's premium service "Uber Black".
• Renault Nissan, is working hard on autonomous vehicles, and unveiled the Next Two at the end of 2015.
• PSA continues to invest in autonomous vehicles and tests their prototypes.
• Tesla wants to be the first manufacturer to market their 100% autonomous vehicles. Their future vehicles should be equipped with fifteen ultrasonic sensors, eight cameras and the Nvidia Drive PX2 on-board computer.
• Toyota is investing one billion dollars in driverless cars and is working with the University of Michigan, Stanford University and the Massachusetts Institute of Technology (MIT) to deploy intelligent functions on its vehicles.
• Volvo continues its efforts in autonomous vehicles and even presented an autonomous truck in a mine.

Here are some scheduled dates for the release of autonomous vehicles:

• 2018: Baidu, Tesla
• 2020: Google, Nissan, Faraday Future, Volvo, Daimler, Honda, Toyota, PSA Group
• 2021: BMW, Ford
• 2030: Hyundai

The automotive industry would shift from an industry of goods to a service industry, knowing that the vehicles would remain the property of the transportation company.

Autonomous cars are also a good means of transport for people with reduced mobility (age, disability) who can ask their vehicle to drive them from point A to point B. The development of autonomous vehicles could even encourage the development of robots taxis, that is to say, a fleet of autonomous vehicles. The automotive industry would shift from an industry of goods to a service industry, knowing that the vehicles would remain the property of the transportation company. The autonomous cars will also be connected and will have an internet connection. They will inform the user and his passengers about the weather, road traffic and the formation of a traffic jam. They will also give passengers time, which will be an opportunity for Google and Apple to introduce embedded services like infotainment in autonomous cars.

However, the arrival of autonomous cars will pose an ethical dilemma. Even though autonomous vehicles will reduce pollution, streamline traffic and avoid 90% of road accidents, it will be necessary to ask how autonomous vehicles will be programmed when they have to choose between saving the life of the passenger or the life of pedestrians?

To illustrate this ethical problem, I will describe the following situation:

An autonomous car with a passenger rolls on the road, and then suddenly a dozen pedestrians appear at the last moment to cross the road, making an inevitable accident. The autonomous car does not have time to brake, so will have to make a choice and this choice is the ethical dilemma of autonomous cars. The autonomous car will have two choices, saving the lives of pedestrians by sacrificing the passenger's life by crashing into a wall or saving the passenger's life by continuing his way crushing pedestrians.

The ethical dilemma of the autonomous car thus leads to a paradoxical situation: with software programmed to save the greatest number, the adoption of the technology of autonomous driving will be slowed down while its objective is to reduce the number of accidents.

Morality would like to save as many people as possible, which implies that the purchased car can kill the passenger. With this information, many people will not want to buy an autonomous car. The ethical dilemma of the autonomous car thus leads to a paradoxical situation: with software programmed to save the greatest number, the adoption of the technology of autonomous driving will be slowed down while its objective is to reduce the number of accidents.

At present, car manufacturers do not yet pronounce on this ethical dilemma but the tendency of the manufacturers is to privilege the safety of the passengers in the autonomous car. For example, Mercedes has decided to prioritize the life of the passengers, the software will be programmed in such a way that passenger survival is priority number one. There is currently no legislation on this ethical dilemma, no instructions on the morals of autonomous cars.

I think that in the future, the management of these ethical problems will have to be solved between three groups who are the customers who buy these autonomous cars, the car manufacturers who develop the software and the government which will be able to regulate the different algorithms that the manufacturers will propose and that consumers will be able to choose.

To put in a nutshell, autonomous vehicles will revolutionize the transport sector and mobility on a daily basis. It is difficult to know when autonomous cars will incorporate our lifestyles, but I think that in two decades, autonomous cars will become part of our daily lives.