A Trip to the Beginning!

Self Driving technology has come a long way from its launch in the late 1900s. In its initial development, Self-driving vehicles depended on a pair of cameras mounted on the sides of the vehicle and street markers identified by the cameras and processed by an analog computer.

The development of this technology, as many may believe – isn’t a phenomenon of today’s advanced technology. Since a major stride in the technology in 1985 and 1986; projects funded by the Defense Advanced Research Projects Agency (DARPA) of the United States and Car manufacturers like Mercedes-Benz and Autoliv inc. (ALV), more-and-more features have gradually been added or improved.

The ALV, in 1985, successfully carried out its self-driving tests on two-lane roads at speeds of 31 Kilometers per hour or 19 mph. It was not until 1986 that the feature of obstacle avoidance became functional. By 1987, the ALV demonstrated that the self-driving vehicles could not only “move” on the road, but they could also now travel in Nighttime conditions as well.

However, the revolution of Self-driving technology began in 1995. Carnegie Mellon’s Navlab 5 (a laboratory designated to the research of autonomous and semi-autonomous vehicles) had demonstrated for the first time ever – an autonomous vehicle test drive that spanned the coasts of the United States.

The test vehicle traveled for 4,501 KM between several states of the US and at a speed of 102.7 km/h.

Since then, Self-driving technology has come a long way.

What are Self-Driving Cars

A self-driving car (also known as an autonomous car or driverless car) is a vehicle that travels between locations without the assistance of a human operator by utilizing sensors, cameras, radar, and artificial intelligence (AI).

To be considered completely autonomous, a vehicle must be able to go to a predefined location without human intervention on roads that have not been modified for its usage.

Audi, BMW, Ford, Google, General Motors, Tesla, Volkswagen, and Volvo are among the companies developing and/or testing self-driving automobiles.

This definition, however, may become blurry when the Devil gets into details. So we would have to make a few clarifications along the line because being automated or autonomous obscures what it means to be self-driving.

Distinguishing Self-driving car Technology Terminologies

A self-driving vehicle should be able to drive itself without the help of a human. In-order words, it should possess its own will. Considering that a vehicle is a machine without its own will, wouldn’t that defeat the purpose of self-driving cars? Well, to solve this dilemma of the technology, for ethical reasons, has been grouped into three broad categories.

Autonomous vs. Automated

Although the terms are often used interchangeably, Scientists like to make a clear and precise definition in order to avoid ethical problems. With Self-driving cars running on AI, it becomes paramount to clearly define the extent of self-control that a car would have.

While all categories of Self-driving cars are capable of driving on their own without human intervention, there is a feature that distinguishes the terminologies associated with the level of “Self-driving”.

Automated Self-driving technology

In Automated Self-driving technology, vehicles are capable of driving on their own. They can read the environment with the aid of sensors, cameras, light detectors, and a myriad of technologies. However, their destination is determined by the passenger of the vehicle.

In philosophical terms, the Vehicle lacks a “mind” of its own and only does the will of its controller. It turns when asked to, stops when asked to, and operates completely on instruction from the driver or passenger.

Autonomous/Self-driving technology

On the other hand, Autonomous/Self-driving technology offers a twist to the plot. To be Autonomous would mean to have no influence whatsoever. But this would imply that a car has a “mind” of its own.

One thing is for certain, autonomous vehicles depend on Machine learning, tons of image processing sensors, and of course – recognition software.

Autonomous control entails adequate performance in the face of large environmental uncertainties, as well as the capacity to correct system faults without the need for external intervention.

To put it simply, an autonomous vehicle, even if directed to travel in North direction, would change this course when it senses a problem or obstacle. But wait, don’t all Self-driving technologies do that? Yes, they do, but there’s more to it. I will explain it below.

Cooperative self-driving Cars

Self-driving technologies aren’t all “autonomous” as one might think. When a self-driving vehicle is not under the influence of another entity, man or machine, then it could be classified as an automated or autonomous vehicle. However, what if the vehicle receives subsequent information from an outside source that directs its path of travel?

Scientists and engineers have had trouble distinguishing all these categories but have come to the conclusion;

Some driving automation systems may indeed be autonomous if they perform all of their functions independently and self-sufficiently, but if they depend on communication and/or cooperation with outside entities, they should be considered cooperative rather than autonomous. – Society of Automotive Engineers, article J3016

Are Electric Cars Self-driving Cars?

Electric cars don’t necessarily have to be Self-driving. Any vehicle could be self-driving, whether it has a battery source of power or fuel. However, that is not to say that electric cars don’t have a certain level of auto-driving technology installed.

Many electric cars have a range of self-driving features like auto-packing, obstacle detection, light sensors, traffic light sensors, speed gauge systems, and much more.

Although the levels of autonomy in recent vehicles are low, some electric cars can drive themselves in ideal conditions.

For example, the Toyota RAV4 is equipped with the Toyota Safety Sense 2.0 suite that includes;

• A Pre-Collision System with Pedestrian Detection.
• Full-Speed Range Dynamic Radar Cruise Control.
• Automatic High Beams.
• Lane Departure Alert with Steering Assist.
• Lane Tracing Assist.
• And Road Sign Assist.

Another top contender with the most investments in self-driving tech is the Tesla Model 3. It comes loaded with full self-driving technology. But even with all its incredible spec, it still isn’t a fully autonomous vehicle.

How Do Self-driving Cars Work

As a car driver, a self-driving car should be able to read its environment and make adjustments or appropriate decisions. To get information from the environment, these vehicles are equipped with an array of advanced techs like image recognition, machine learning, and neural networks.

The image recognition systems collect images from the road, like lights, trees, curbs, street signs, and much more. Collecting the data is just the easiest part, the neural networks handle the more difficult task.

The neural networks identify patterns in the image data and send this data to the machine learning algorithms to process them and retain that information. The algorithms use the data collected to analyze a situation and take the most viable route or take the best “decision.”

Although this system still needs lots of improvement, you’d be astonished at how advanced self-driving vehicles have become.

Some Electric Cars with Self-driving Technology

Regarding vehicles that can be fully automated, Tesla takes the crown. However, there a tons of other vehicles that have automated driving capabilities.

• Acura: TLX
• Alfa Romeo: Giulia, Stelvio
• Audi: A4, A5, A6, A7, A8, E-Tron, Q3, Q5, Q7, Q8
• BMW: X3, X4
• Ford: Bronco Sport, Edge, Escape, Explorer
• Genesis: G80, G90, GV80
• Hyundai: Elantra, Ioniq, Nexo, Palisade, Santa Fe, Sonata
• Infiniti: QX50
• Jeep: Grand Cherokee L
• Kia: K5, Niro EV, Seltos, Sorento, Sportage, Telluride
• Land Rover: Range Rover, Range Rover Sport
• Lexus: ES, IS, LC, LS, NX, RX, UX
• Lincoln: Aviator, Corsair, Nautilus
• Maserati: Ghibli, Levante, Quattroporte
• Mercedes-Benz: AMG-GT 4-Door, A-Class, C-Class, CLA-Class, CLS-Class, E-Class, GLA-Class, GLB-Class, GLC-Class, GLE-Class, GLS-Class, S-Class, SL-Class
• Nissan: Altima, Leaf, Rogue, Rogue Sport
• Polestar: Polestar 1, Polestar 2
• Porsche: Cayenne, Taycan
• Subaru: Ascent, Crosstrek, Forester, Legacy, Outback
• Tesla Model 3
• Tesla Model S
• Tesla Model X
• Tesla Model Y
• Toyota: Camry, C-HR, Corolla, Highlander, Mirai, Prius, RAV4, Sienna, Venza
• Volkswagen: Arteon, Atlas, Atlas Cross Sport, ID.4
• Volvo: S60, S90, V60, V90, XC40, XC60, XC90

Will cars ever be fully autonomous?

Presently, there are no fully autonomous vehicles available to the public. There are a lot of criteria to be met, technology-wise, economy-wise and social problems to be solved be vehicles can be fully autonomous.

Artificial Intelligence Problem

Artificial intelligence is becoming smarter every day, but it’s grossly lacking in making effective decisions. One of the many fears of relying on current AI is the fact that the AI is not perfect and may misread the environment. For example, the AI might misidentify an obstacle, a pedestrian, or even a cycler.

If you’re a driver, you know that driving is not just about turning the wheels; it also comprises reading cues from other vehicles, drivers, and environment conditions. In a bustling environment, the AI may crash, might be incapable of identifying all obstacles, or might miss out on a few.

You have to be precise when it comes to driving because one wrong turn could be catastrophic.

Security Concerns

All systems have a loop in them. Electric self-driving vehicles rely on computer software to function. This means that it could be compromised and controlled by a malicious entity.

Any system operating with an OS is open to cyber security threats. Malicious entities may use this to obtain your travel destinations, lock you in the car, cause the vehicle to forcefully change travel directions, or cause a malfunction in the system.

The security threat is one not taken lightly by engineers and government agencies.

Environmental Conditions

The vehicle may not perform well in varying weather conditions. The sensors which the AI depends on to drive itself may become useless, blind, or malfunction when under harsh weather conditions like a rainstorm.

As the AI is heavily dependent on photographic images, this may be unreliable during a snowstorm, etc. The AI may also not be able to adjust the speed the direction of the electric car under these conditions, making self-driving a difficult task.

Road Infrastructure

Would the AI of automated cars be able to detect good roads from bad roads? Would it be able to identify faulty traffic lights? How about reading unclear road signs? These are critical aspects of automated driving that need to be put into check before any society would allow autonomous cars to travel any busy highway.

Political Acceptance

Even though Electric cars may prove to be capable of cutting down urban carbon emissions to 80%, they may still come under heavy scrutiny by politics and society in general, as people would also be skeptical about handing over their safety to an “automobile” (pun intended).

Moral Problems

Who takes responsibility for accidents that are caused by autonomous vehicles? This poses a serious ethical problem hindering the progress of self-driving vehicles. Is the owner of an automated vehicle responsible for any damages caused by an autonomous car? What if such accidents result from technical failures? Would the producer of the vehicle be held accountable for the damage done?

This is quite the dilemma as companies, and car owners may battle to exclude themselves from the blame of automated driving.

Another factor to consider when discussing the autonomous capability of electric cars is the Levels of Automation possible in Self-Driving Cars.

Autonomous technologies are categorized into different levels, with the subsequent levels having the more advanced features with the capability of having more independent control over the driving process.

• Level 0: At this level, the automated system has no sustained control. It may, however, issue warnings and, for some time, intervene during driving. The Passenger has more control over the vehicle.
• Level 1: In this mode -also called “hands-on”-control over the vehicle is interchangeable between the automated system and the driver. At this level, the automated system may be able to set speed, adjust engine power, brake power, and also offer parking assistance. However, the driver should be on the lookout.
• Level 2: The AI takes over the vehicle completely, controlling brakes acceleration and steering.
• Level 3: The driver may take his eyes off the whole driving process in this mode. The driver can engage in other activities while the automated system takes over the driving process. However, the driver must still be prepared to take over within a certain period of time or in a more chaotic driving environment.
• Level 4: The human can relax with ease of mind. Hence, this mode is called “mind off.” In this case, the driver may sleep or leave the driver’s seat. Although, this hasn’t been cleared for rowdy highways yet.
• Level 5: This level is the dream of all automated vehicles technology. No human supervision would be required to operate the vehicle.

Bearing this in mind, the owner would have to know if his vehicle can drive on its own.

Conclusion

While there has been remarkable progress over the years, there is still a milestone to go in the world of self-driving technology. Different grades of cars can travel on highways, and some can’t. It’s up to the owner to determine what level of automation his vehicle possesses.