PDF version of this report
You must have Adobe Acrobat reader to view, save, or print PDF files. The
reader is available for free
download.

Self-Driving Vehicles

by Michael Gariffo

Copyright 2022, Faulkner Information Services. All
Rights Reserved.

Docid: 00021048

Publication Date: 2209

Report Type: TUTORIAL

Preview

In the march toward trying to turn science fiction technology into
reality, one of humanity’s latest projects is the self-driving vehicle.
Having been a dream of mankind since before the automobile even existed,
self-driving vehicles could provide massive benefits to the world’s
populace. Not only could they offer additional work and leisure time for
billions of people around the globe struggling through their daily
commute, but they also have the potential to save hundreds of thousands of
lives by eliminating the human errors that cause nearly all traffic
accidents. Unfortunately for our sci-fi dreams, there is a long road ahead
for the technologies, legislation, and infrastructure that will be needed
to make self-driving vehicles a part of peoples’ day-to-day lives. This
report will examine the current definitions of self-driving vehicles, take
a look at the technologies powering them, and examine the legislation and
public reaction contributing to the development of this new type of
automated transportation.

Report Contents:

• Executive Summary
• Description
• Current View
• Outlook
• Summary
• Web Links
• Related Reports

Executive Summary

[return to top of this
report]

Self-driving vehicles are currently one of the hottest areas of
technological development. This is due primarily to the potential they
could provide to revolutionize not just day-to-day travel, but also for
shipping, traffic management, and road safety. The technology has been
dreamed of for decades but is only now being made possible through a
combination of improved sensors, computers, and networking developments.
These developments combine to create a vehicle that is very close to being
able to replace a human driver’s decision-making capabilities and
responsiveness. It is the remaining gap between human abilities and
machine prowess that stands as the last obstacle for the widespread
commercialization of self-driving vehicles.

While early attempts at semi-autonomous cars are already on the market,
no company has yet to publicly offer a truly autonomous vehicle. This is
due to limitations in current technology and the deep concerns of
lawmakers, engineers, and the public at large to the idea of handing the
safety of the driving public over to a machine. To address these concerns,
automakers and developers at several companies are working feverishly to
be the first to bring a commercially viable, fully autonomous car to
market. While the first of these will likely be met with some trepidation,
it will not be until the public gets its hands on the initial round of
self-driving vehicles that they will have any chance of coming to be
accepted as a safe, viable mode of transportation and shipping.

To this end, companies like Google (through its Waymo subsidiary), Tesla,
Mercedes-Benz Group (formerly Daimler), Apple, and others are now
strengthening their technological holdings, track records, and development
teams in order to meet the restrictions that will be placed on
self-driving vehicles by federal and state legislators. Since the concept
of a self-driving car or truck was so far-fetched for many years, no laws
were on the books to govern their operation. However, this is no longer
the case, as the US Department of Transportation (US DOT) has now released
and updated a set of federal
guidelines that are expected to lay the groundwork for
municipalities and agencies across the country, deciding how they govern
the operation of automated vehicles.

Description

[return to top of this
report]

What Is a Self-Driving Vehicle?

To understand this topic it is important to first establish a strict
definition of what exactly a self-driving vehicle is. This may seem simple
on the surface: a self-driving vehicle is simply a car or truck that
drives itself instead of a human doing it. However, there are many
gradations of autonomy between automobiles that depend entirely on the
driver and those that could complete an entire journey with no human
intervention whatsoever. Things like parking assist, adaptive cruise
control, lane guidance warnings, automatic braking, and other safety and
convenience technologies have been available in some automobile lines for
several years. Are these self-driving vehicles? No, but they are getting
increasingly close. To better understand exactly where past and current
vehicles fall on this spectrum, the SAE, an industry group made up of
numerous automakers and designers, published its levels of automobile
autonomy in 2014. This list focuses on how ready-to-take-control of the
vehicle the driver must be on each level, topping out at the lofty goal of
a vehicle operating entirely without human interaction. Although they had
previously developed their own list of automation levels, the US
Department of Transportation has officially adopted the SAE outline.

• Level 0: The automobile may issue warnings to the driver, but has no
  direct control over itself.
  • This would corresponding to some of the aforementioned
    technologies, such as sensors that detect when a driver has wandered
    from a traffic lane or even backup sensors that detect obstacles in
    the automobile’s path.
• Level 1: A vehicle in which the driver must be ready to take control
  at any time but also has partial control over its own movements.
  • This includes technologies such as automated parking controls,
    adaptive cruise control, and other features that allow the car to
    control itself in a very limited fashion and only within certain
    circumstances.
• Level 2: The vehicle handles its own acceleration, braking, and
  steering, but still requires the presence of a human driver to take over
  at a moment’s notice.
  • This is essentially the top level of what is currently available
    commercially, with the only exception being Tesla’s Autopilot
    feature making it to this level of autonomy.
• Level 3: Drivers can completely turn over control of the vehicle to
  the onboard systems but only within predictable, controlled environments
  such as limited access freeways.
  • This refers to a scenario in which a self-driving car has advanced
    to the point where it can reliably handle itself in the generally
    predictable traffic situations found on most freeways or highways
    but has not sufficiently advanced to the point where the more
    complicated aspects of city driving can be handled.
• Level 4: The vehicle is entirely automated in nearly all situations
  with a few scenarios (such as ongoing extreme weather) requiring a human
  driver to take over.
  • On this level, self-driving cars begin to become a reality with
    human intervention being needed only when a situation becomes
    unpredictable such as in the event of a sudden rainstorm or
    reduction of visibility.
• Level 5: Full automation, from origination point to destination with
  no human input whatsoever.
  • This is the highest level of autonomy with cars having replaced
    the human driver entirely. It requires an onboard system capable of
    driving anywhere and in any conditions that it are currently legal
    for a human.¹ 

These classifications of automation are extremely useful for future
development, but they come at a time when technology is still just
scratching the surface of what is possible. The SAE grades currently
available experimental models at level 3, possibly 4 in a few cases, with
commercial models definitely ranking no higher than level 3. Numerically
speaking, there is only a one-to-two digit difference between what is
possible now and the dream of full automation. However, those one or two
ranks represents a chasm of real-world difference that engineers and
automakers must cross.

As things stand, only a few vehicles even attempt full automation. These
are nearly all experimental models that are only driven under tightly
controller circumstances with human drivers always on board to take
control in an instant. The lives of everyone onboard an automated vehicle,
as well as anyone in the vicinity, are quite literally in the hands of the
onboard systems. Because of this, legislators and automakers must have
ironclad confidence in the systems’ abilities to make the best possible
decisions in essentially 100 percent of all scenarios before the top level
becomes a commercial possibility.

While this examination may seem pessimistic, it is quite the opposite. We
are, in many ways, very close to achieving the necessary levels of
technological advancement needed to make the leap of confidence possible.
It may still require years of testing to verify that full automation is
possible, but essentially all of the building blocks are already in place
to create a fully automated vehicle. The next section will go over the
technologies currently being used by experimental automated vehicles,
including explanations of how they function and the role they play in
vehicle operations.

Self-Driving Vehicle Technologies

In order for self-driving vehicles to operate, they must essentially
replace a human driver. To fulfill this goal, they need to mirror the
driver’s ability to assess a given situation, analyze available options,
and take the best course of action. Humans do this by perceiving the road
with their eyes and ears, using their brains to analyze that input, and
then physically enacting their decisions upon the vehicle to take the best
course of action. For an automated vehicle, the process takes several more
steps and much, much more hardware. That is not to say there are not
similarities; where a human has eyes, a self-driving vehicle would have
various forms of sensors; where a human has a brain, the vehicle would
have a computer constantly analyzing current conditions and planning
courses of action. Finally, the human driver’s muscles would be replaced
with a series of motors and actuators to actually control the
vehicle. 

Figure 1. Waymo Self-Driving Car Prototype

Source: Waymo

Radar

The same technology created to detect ships at sea and planes in the air
is an integral part of the spatial awareness in modern self-driving
vehicles. As with the previously mentioned applications, radar in
self-driving vehicles uses radio waves to determine the location, shape,
and motion of objects surrounding its origination point. On a basic level,
radar accomplishes this by sending out a radio wave and detecting when the
return wave has bounced off an object and made its way back to the
origination. This tells the radar sensors that there is an obstruction,
and provides information such as the obstruction’s location, direction of
travel, velocity, and more. This makes the technology ideal for
determining factors like the presence of other cars, pedestrians, road
obstacles, and any other potential danger or factor that a human driver
would process into their decisions as a matter of course. Radar does have
its limitations, however. It does not have the resolution of some of the
other technologies that will be discussed here, meaning that it cannot, on
its own, always determine exactly what shape an object is. Certain factors
can also interfere with the radio waves including large obstructions,
materials capable of absorbing the waves, extreme weather, and other
factors. That said, radar is still one of the most important tools in the
arsenal of self-driving vehicles today and one that is being constantly
improved upon.

Lidar

Lidar is, in many ways, radar’s lesser-known cousin. Lidar (which was
literally coined as a combination of “light” and “radar”) also detects
nearby objects by bouncing a signal off them and monitoring its return.
However, instead of radio waves, it uses light, typically in the form of
laser light. While this technology does not have the same ability to
penetrate fog or sense long distances through rain as radar, it does have
a significantly higher resolution allowing self-driving vehicles to
differentiate between, for example, a person and a person-sized tree.
Where radar might miss some of the surface characteristics of this object,
seeing only that it was about six feet tall and one and a half feet wide,
lidar would be able to assess the silhouette of a human being, giving the
computer the necessary information to predict the motion and future
placement of that person. These two technologies combine to create a
single field of view that allows for the early detection of distant
objects that may require avoidance (via radar) as well as the specific
recognition of those objects as they approach the vehicle (via
lidar). 

Figure 2. A Visual Representation of Sensor Feedback from
Google’s Self-Driving Car

Source: Google

Computer Vision

This is the third component to the visual sensory capabilities of a
self-driving vehicle – the one that makes it possible for the automobile
to take the acquired information and process it into actionable data. The
term “computer vision” refers to a field of technology that focuses on
allowing computers to obtain and process visual information with the
ultimate goal of utilizing the data to determine and zero in on the best
possible decision in a given situation.² The greatest area of
interest for this technology is its ability to allow computers to perform
tasks that would typically require the processing capabilities of a
human’s eyes and brain. To better illustrate how this may be useful, look
at the above image. The radar, lidar, and other visual sensors in Google’s
self-driving car have captured the various obstacles (in green) that must
be taken into account when planning a route. That raw data is processed by
the vehicle’s computer vision system to classify the type of object each
detected obstacle is. Notice the black lines around what appear to be
parked cars or other stationary objects and the pink boxes surrounding
what appear to be other cars driving on the same road. It is the car’s
computer vision that allows it to make this differentiation, giving its
systems the information necessary to predict possible outcomes. For
instance, the car would know to monitor the other vehicle on the opposing
side of the intersection for movement as it needs to make a left turn that
will cross that vehicle’s path. Knowing this, it can determine a course of
action, which, in this case, would likely be to wait for that car to cross
the intersection before completing its left turn. In a sense, computer
vision provides the self-driving vehicle’s vision processing capabilities,
passing its processed information along to the systems that use the info
to make the same decisions a driver must make every second they are behind
the wheel. 

Odometry

This is perhaps the simplest technology in use as part of the
self-driving vehicle’s systems. Odometry refers to the monitoring of an
object’s motion in order to determine its location. By using this
technology, a self-driving vehicle can take its starting position, apply
the amount of forward progress it has made, and determine its new
positions by simply doing the math with the amount of distance traveled.
It should be noted that, in order for odometry to be effective, the
vehicle must have a very accurate idea of its starting point. In most
cases, this initial calibration is made possible by the last major
technological component of self-driving vehicle’s sensor systems, GPS.

GPS

Although its level of precision in determining the exact location of a
self-driving vehicle can be less than perfect at times, GPS is an integral
part of self-driving vehicles. Essentially, it fulfills two major roles:
it lets a car knows where its starting point is, making the use of
odometry possible; and it tells the vehicle where the roads are. While
this second purpose may seem obvious, it is nonetheless important as it
forms the basis for how the self-driving vehicle begins planning its route
and how it executes that route. It is important to note that even the most
accurate publicly available GPS systems today can only provide a position
down to within a few feet. Thankfully, this is more than enough accuracy
to serve as the aforementioned basis for the vehicle to determine its
exact position. In order to tighten its tolerance to a point where the
vehicle knows exactly where it is, the onboard computer
takes the GPS data and subsequently applies the information being derived
from the radar, lidar, and computer vision systems. This massive amount of
data is then processed to form a single determination on the car’s
location, speed, and its position in relation to other vehicles and
obstacles.

Summary

The level of complexity provided by the technological systems mentioned
here should begin to reveal the level of advanced sensing and processing
required to replicate the human driver’s ability to operate a vehicle.
Although, in some ways, these systems have much, much more information
available to them than any human driver ever could, they also must
collect, process, and act upon this information using their very non-human
systems. Computers’ ability to mimic the behaviors of human beings has
always been a sticking point of artificial intelligence development.
Thankfully, we are closer than ever to the point where an advanced sensor
array and computer system can indeed replace a human being, at least in
this one capacity. 

Current View

[return to top of this
report]

Legislation

Now that the technological aspects of what makes self-driving cars
possible has been made clear, the logical next step is to examine how
these vehicles might be able to operate in the real world. In order for
this to happen, they must, of course, satisfy the same governmental
guidelines that any human-operated vehicle needs to fulfill. However, the
unusual nature of the vehicle’s operator also requires an additional set
of legislation to assure that the self-driving vehicles, at worst, pose no
greater threat to their occupants or fellow vehicles than the average
human driver would. While it is still early days for this kind of
legislation, several states have already taken a stance on the matter.
More importantly, the federal government has now weighed in on
self-driving vehicles for the first time in the form of a set of
guidelines published by the US Department of Transportation (DOT). This
section will examine the various forms of legislation surrounding
self-driving vehicles that are already on the books, as well as
speculating on what new laws these vehicles may cause to be created.

State Laws

As of 2020, a total of 41 states have officially enacted some form of
legislation or executive order governing the operation of self-driving
vehicles. This represents a massive uptick from the 2017, with 20 of those
having put their first laws of this nature into the books within the
previous 36 months.³ Although a wide range of states have
addressed the upcoming need for laws governing autonomous vehicles, only a
handful have actually included language that allows for the testing of
self-driving cars. However, governing bodies like the National Conference
of State Legislatures (NCSL) have recognized this rapidly changing legal
situation, and have created tools like the
Autonomous Vehicle Legislative Database. This online system allows
lawmakers and citizens alike to track the current laws and proposed bills
that have been passed or are under current consideration for their state
and other states around the country.

While many of the aforementioned laws were put in place to prevent
pre-emptive introductions of unsafe vehicles, many also show a promising
tendency by US lawmakers to encourage the continued development of
self-driving technology, and a willingness to allow the technology to take
that final step in joining human drivers on the road. Nowhere is this more
true than in Nevada, which became the first jurisdiction in the world to
specifically allow the operation of self-driving vehicles on its roads in
2011.⁴ That legislation officially allowed the Nevada
Department of Motor Vehicles (NDMV) to begin setting safety standards and
guidelines for the operation of these types of vehicles within states
borders.⁵ Unsurprisingly, Google fully supported the law, as it
was just then beginning to approach public road tests for its early
self-driving vehicle prototypes.⁶ The Nevada law would prove to
be a prototype for other states, setting a precedent for factors such as
requirements that a human driver be onboard at all times.⁷

While Nevada may have been one of the most progressive states in terms of
giving this fledgling technology a chance, California appears to be poised
become one of the most liberal jurisdictions in the world for the types of
self-driving vehicles that are legally allowed to operate on its roads.
Unlike Nevada, California has passed legislation that allows the operation
of autonomous taxis that are completely unmanned.⁸ While this
is still a limited-scope pilot program, it does show a willingness to be
on the bleeding edge of self-driving car testing.

Arizona became a notable participant in the self-driving vehicle
revolution when it was revealed that UPS had been testing a fleet of
self-driving trucks through a 100-mile route between Phoenix and Tucson.⁹
The trucks, which supposedly operate at “level 4” on the SAE scale, had
been in operation on the roads without incident for several months before
their existence even became public knowledge, suggesting they had been
operating relatively flawlessly.¹⁰

Federal Laws

State laws on the matter are very important; however, the looming threat
of federal regulations and restrictions was a long-term worry for many
self-driving car developers and manufacturers. Thankfully, the US
government has finally clarified its stance on the new area of technology
in the form of a set of guidelines published by the US Department of
Transportation. The so-called “Federal Automated Vehicles Policy” outlines
a set of 15 safety assessment objectives that self-driving car makers must
assure that their vehicles meet in order to be legally allowed on the
road.^11,12 These guidelines focus on areas such as how
operational data must be recorded and shared, how vehicles are to behave
following a crash, standards for user interfaces, and other safety and
privacy guidelines. The document also provides strict delineation between
which regulatory responsibilities will be assigned to federal and state
agencies going forward. 

While the federal government has, largely shown positive movement in the
eventual adoption of self-driving technology on a national level. It has
faced opposition from some groups that may not be obvious opponents to the
proliferation of automated vehicles. Among these groups are unions
involving transportation and transit workers.¹³ These
organizations’ members would stand to lose their jobs if they could
potentially be replaced by a self-driving delivery or transport vehicle,
making their opposition entirely understandable.

It seems that, like the aforementioned state legislatures, the federal
government’s current stance towards self-driving car technology is a
friendly one, but also one that will face strong pushback from certain
groups with a vested interest in maintaining the status quo.

Outlook

[return to top of this
report]

Benefits

Although the potential benefits of self-driving technology are likely
obvious to most, it is still important to look at the specific benefits
and potentially world-changing impact that the proliferation of such a
technology could bring with it. Below is a list of these benefits:

• Cost Savings – A study conducted by Morgan Stanley
  determined that the total reduction in costs for the US economy, should
  self-driving cars become commonplace, would be in the range of $1.3
  trillion per year. These savings would be derived from a combination in
  the reduction of costs associated with traffic accidents ($563 billion
  per year), a reduction in fuel consumption ($158 billion per year),
  reduced congestion ($138 billion per year), and the elimination of
  productivity loss due to being stuck in traffic ($422 billion per year).¹⁴
• Productivity – This point obviously ties heavily
  into the previous one. The same Morgan Stanley study estimates that
  drivers in the US spend a total of 75 billion hours per year on daily
  commutes. Assuming that people in their self-driving vehicles would be
  90 percent as productive as those at their desks, the report estimates
  that the aforementioned $422 billion per year could be generated by a
  gain of $25 per hour of returned productivity due to work being able to
  get done during travel. Similarly, less business-oriented trips could
  also be made more pleasant by allowing families to enjoy leisure time
  together during a road trip, or allowing tired drivers to rest before
  reaching their destination.¹⁵
• Safety – Again, the previous point leads into the
  next, this time focusing on the potential safety benefits. The Morgan
  Stanley report pegs the portion of vehicular deaths in a given year that
  are caused by human error at 90 percent of the 32,385 annual fatalities
  reported in the US.¹⁶ This means nearly 30,000 lives per year
  could be saved by eliminating human fallibility from the equation. This
  aspect of self-driving vehicles would be particularly beneficial in
  curtailing negative factors in driving safety such as tired driving,
  driving under the influence, and distracted driving. 
• Transport of Goods – Although the majority of
  companies working on the technology are focused on developing
  self-driving passenger vehicles, there is indisputably a market for
  autonomous delivery vehicles. Imagine the possibility of the delivery
  drones currently being developed by the likes of Google and Amazon
  combined with one of Mercedes-Benz’s self-driving tractor trailers. The
  trailer could be motoring down the highway with thousand of deliveries
  on board, disgorging a fleet of drones as it went to deliver those
  packages. Thanks to the machine-to-machine network this scenario would
  include, those drones could make deliveries much further from a
  distribution center than would otherwise be possible while the truck
  could avoid hours worth of stops on any given trip. 

These are just a fraction of the possible benefits that self-driving cars
and trucks could bring, with the possibilities for even more just now
being realized. 

Drawbacks

Like any revolutionary technology, self-driving vehicles bring with them
certain concerns. Thankfully, most of these are relatively minor,
impacting only a small percentage of the population or being easily
ameliorated by careful regulation and testing. They include:

• Safety – This is the most obvious one, and the first
  thing likely to make most people hesitant when stepping into their first
  self-driving car. Thankfully, as mentioned above, self-driving cars are
  already achieving safety records far exceeding those of the most
  experienced, safest human drivers in the world. This does not guarantee
  that new problems or issues will not arise as the proliferation of
  self-driving vehicles reaches a massive scale. However, early
  governmental intervention, such as the Department of Transportation’s 15
  guidelines, are already ensuring that future self-driving cars will be
  even safer than what is currently available. 
• Job Loss – The transportation of goods and people
  provides jobs for many tens of thousands of people in the US alone.
  Although a future where these workers have been entirely replaced by
  self-driving vehicles is likely still decades off, the threat to their
  job security is none the less real. However, this is really no different
  than the continued mechanization and automation of the modern factory,
  and it provides no greater threat to the stability of the economy as a
  whole than any other continued development of technology. 

Public Reception

We’ve covered the commercial interests involved in the development of
self-driving vehicles, as well as the governmental and regulatory
responses. Now we take a look at the general public’s perception of the
concept of being driven around by an artificially intelligent car rather
than their own hands and feet. 

As mentioned several times in this report, self-driving cars can be
disconcerting. While many science-fiction movies have used the trope of an
automated vehicle, many horror movies have also taken the concept to its
most fearful extreme. Films like “Maximum Overdrive” and stories like
“Christine” provide us with examples of malicious automobiles out to kill
as many humans as possible, while positive examples like “Knight Rider”
and the “Transformers” movies and cartoons provide us with examples of
friendly autonomous cars out to protect us and keep us safe. Although no
one in their right mind would consider any of these works of fiction to be
accurate predictors of our future, the act of examining a culture’s
fiction surrounding a given topic can, none the less, prove revelatory
towards that culture’s feelings on their central conceit becoming a
reality. As with extraterrestrials, flying cars, and various other sci-fi
predictions, the public reaction seems as if it would generally be mixed,
with an overall preponderance towards apprehension. 

Thankfully, more concrete representations of the public’s desire for
self-driving vehicles have also been captured. One of the earliest of
these was a survey conducted by Accenture in 2011, in which 49 percent of
respondents in the UK, just barely a minority, claimed they would be
comfortable in a self-driving vehicle.¹⁷ US drivers appear to
be more apprehensive, as a 2012 JD Power and Associates survey found only
37 percent would be interested in purchasing an autonomous car for
themselves. This figure dropped to just 20 percent when the same
respondents were told this extra technology would add $3,000 to the
purchase price of the vehicle.¹⁸ To be fair, self-driving cars
were even more of an unknown quantity when these surveys were
conducted. 

A 2014 study conducted by Insurance.com found that three quarters of
those surveyed would be interested in purchasing a self-driving car, with
that figure rising to 86 percent if the vehicles were cheaper than their
manual counterparts. Meanwhile, 31.7 percent said they would continue to
use manual vehicles no matter what the availability of autonomous ones
turned out to be.¹⁹

Unfortunately, incidents like Google’s first at-fault accident and fatal
crashes from Tesla and Uber may weigh on the minds of potential customers.
However, the overall safety statistics of the early attempts at a
self-driving vehicle remain impressive. That said, this does not mean that
all, or even most people would be swayed towards purchasing an autonomous
vehicle for themselves. Believing that a self-driving car should be
allowed on the road and putting the lives of one’s own family on the line
in such a vehicle are two very different things. It will very likely not
be until the first of these fully automated vehicles reaches the market
and the possibility of owning one becomes a reality that it will become
truly apparent how willing humans are to essentially enlist a robot
butler. 

What is already clear is that a big portion of this decision will
apparently come down to the cost of these autonomous vehicles when
compared to manual ones. Two of the aforementioned surveys showed that
price was a major factor in determining which type of vehicle to buy.
Unfortunately for the early producers of self-driving vehicles, it is
very, very unlikely that they will be able to produce a commercially
available product that is capable of full autonomy for anything near what
a manual car can be sold for. For example, Tesla’s Model S, which is only
semi-autonomous, currently has a sticker price of approximately $105,000+
in the US.²⁰ Although this pales in comparison to the cost of
some “super cars” and higher-end luxury models, it is, none the less,
several times what the most commonly sold vehicles are priced at. As with
any technology, this price gap is likely to shrink as the technology
improves, and as more competitors enter the self-driving vehicle game.
Just as smartphones, televisions, and computers have come down in price
over the years, so too will self-driving vehicles. The question is really
just how long it will take, and what will serve as the breaking point
where these types of cars and trucks stop being playthings for the
technological and financial elite and become daily drivers for the average
family.

Summary

[return to top of this
report]

At the risk of sounding like the same broken record that tech journalists
have been playing for years, self-driving vehicles do indeed remain at
least several years off. While the technology to produce a self-driving
car or truck already exists, the reliability must continue to be
thoroughly vetted, the governmental response to the entirely new form of
travel must continue to develop, and the commercialization of this
technology must continue to be pursued. Early attempts like Tesla’s
Autopilot are excellent predictors for not just how the technology will
develop, but also for how people will respond do it. So far, that response
from the general public seems to be largely positive, but with a strong
vein of completely understandable apprehension. Meanwhile, the US
government has shows ample desire to promote and encourage the adoption of
this technology thanks to the myriad benefits its poses for public and
economic well-being. 

The next several years will prove integral to the future of self-driving
vehicles. While a continually positive safety record will put the
technology on the fast-track to reaching America’s driveways, other
high-profile accidents could just as easily push it back into the realm of
science fiction. It is up to the engineers, automakers, regulators, and
private citizens which of these roads self-driving vehicles travel
down. 

References

¹ “SAE Levels of Driving Automation Refined for Clarity and
International Audience” (blog). SAE. May 3, 2021.

² Carlo, Vandoni E. and Huang, T. “Computer Vision:
Evolution and Promise.” 1996 CERN School of Computing. Retrieved
September 2022.

³ “Autonomous Vehicles | Self-Driving Vehicles Enacted
Legislation.” National Conference of State Legislatures. Retrieved
September 2022.

⁴ “Nevada Enacts Law Authorizing Autonomous (Driverless)
Vehicles.” Green Car Congress. June 25, 2011.

⁵ Ibid.

⁶ Markoff, John. “Google Lobbies Nevada to Allow
Self-Driving Cars.” The New York Times. May 11, 2011.

⁷ Healey, Tim. “Nevada Passes Law Allowing Self-Driving
Car.” Motor Authority. June 24, 2011.

⁸ Austin, Patrick Lucas. “Self-Driving Cars Can Now Pick Up
Passengers in California–But Only for Free Rides.” Gizmodo.
June 1, 2018.

⁹ DeAngelis, Marc. “UPS Is Testing Self-Driving Trucks in
Arizona.” Engadget. August 15, 2019.

¹⁰ Ibid.

¹¹ “Fact Sheet: Federal Automated Vehicles Policy
Overview.” US Department of Transportation. Retrieved September 2022.

¹² “Federal Automated Vehicles Policy.” US Department of
Transportation. Retrieved September 2022.

¹³ Shepardson, David. “US Push for Self-Driving Cars Faces
Union, Lawyers Opposition.” Reuters. June 16, 2021.

¹⁴ Zhang, Benjamin. “Autonomous Cars Could Save the US $1.3
Trillion Dollar a Year” [infographic]. Business Insider.
September 12, 2014.

¹⁵ Ibid.

¹⁶ Ibid.

¹⁷ “Consumers in US and UK Frustrated with Intelligent
Devices That Frequently Crash or Freeze, New Accenture Survey Finds.” Accenture.
February 10, 2011.

¹⁸ Yvkoff, Liane. “Many Car Buyers Show Interest in
Autonomous Car Tech.” Cnet. April 27, 2012.

¹⁹ “Autonomous Cars: Bring ‘em On, Drivers Say in
Insurance.com Survey.” Insurance.com. July 28, 2014.

²⁰ Doll, Scooter. “2022 Tesla Prices: How Much Does Your
Favorite Model Cost?” Electrek. September 14, 2022.

Web Links

National Highway Traffic Safety Administration (NHTSA): https://www.nhtsa.gov/
National Transit Safety Board: https://www.ntsb.gov/
SAE: https://www.sae.org/

About the Author

[return to top of this
report]

Michael Gariffo is an editor for Faulkner Information
Services. He tracks and writes about enterprise software, the Web, and the
IT services sector, as well as telecommunications and data networking.

[return to top of this
report]