Content uploaded by Paul Maxwell
Author content
All content in this area was uploaded by Paul Maxwell on Oct 16, 2018
Content may be subject to copyright.
2018 IEEE International Symposium on Technology in Society (ISTAS) Proceedings
Miriam Cunningham and Paul Cunningham (Eds)
ISBN: 978-1-5386-9479-4
The Unforeseen in Unmanned Vehicles
Paul Maxwell
Army Cyber Institute
West Point, NY USA
Email: paul.maxwell@westpoint.edu
Michael Nowatkowski
Augusta University Cyber Institute
Augusta, GA USA
Email: mnowatkowski@augusta.edu
Abstract - The development of unmanned vehicle technology
is rapidly proceeding and will result in numerous advances in
autonomous vehicles. Most of the research effort to date focuses
on the safe and effective operation of these vehicles that will allow
them to integrate into society. A research gap exists though in
the technical, policy, and legal fields regarding illicit use of these
vehicles beyond their programmed functions. In this paper we
explore possible misuse of unmanned vehicles and illustrate the
need for research in the technical, policy, and legal realms.
Keywords - Unmanned Vehicles, Cyber Security, Cyber Policy,
Cyber Law
I. INTRODUCTION
Unmanned, autonomous vehicles of all types, e.g., air,
ground, water, are rapidly advancing in capability. Plentiful
research is ongoing to solve the challenges of safe operation for
these systems in our complex environment. This research is
necessary for the eventual inclusion of these systems into
society. With that task now well underway, it is time to shift
some research effort into other aspects of unmanned systems
such as security and privacy. Before these systems begin mass
production, it would be wise for society to develop rules and
regulations protecting both public safety and privacy. As we
have seen with other systems, it is more efficient to create
designs from the ground up that provide these protections
instead of attempting to patch them once fielded. Many in
industry would argue to develop then regulate similar to how
the automobile and aviation industries developed. However,
this method often results in long periods of anguish for users as
issues are slowly resolved. Failure to develop solutions to
these issues now will eventually result in the use of unmanned
vehicles (UVs) in ways that are harmful to society.
As new technology is developed, it often provides benefits
to society. It enables people to be more productive, to work
more safely, and to live healthier lives. Frequently though, this
same technology is co-opted for battlefield or criminal use.
Cell phones enable remote triggering of bombs or the illicit
tracking of people [1]. Unmanned Aerial Vehicles (UAVs) are
used to deliver explosives as documented in attacks by ISIS
and the recent assassination attempt in Venezuela. Social
media provides command and control networks and recruiting
tools for Al Qaeda and ISIS terrorists. Internet of Things (IOT)
devices facilitate massive Distributed Denial of Services
Attacks [2]. GPS devices are used to stalk people [3].
Eventually unmanned vehicles will be utilized by criminals or
terrorists for nefarious purposes as well.
Some of the illicit applications for unmanned vehicles may
affect privacy and personal freedoms while others may pose a
direct threat to public safety. Many questions arise from these
possibilities such as how will stalking or harassment by
unmanned vehicles be prevented? Will citizens be protected
from constant surveillance that these technologies enable?
How will law enforcement or the military stop an unmanned
vehicle and how will it gain authorization to search it? How
will the perpetrator of an illegal act using an unmanned vehicle
be attributed? The answers to these questions and many
similar ones are currently unknown. The purpose of this paper
is to propose the start of the discussion on these topics
including potential technical, legal, and policy solutions to
these issues so that the solutions are emplaced prior to mass
adoption of these technologies.
To initiate the discussion, the remainder of the paper is
organized as follows. Section 2 discusses work related to
unmanned vehicles and public safety/privacy. In section 3, we
will use several scenarios to illustrate the gaps in current
research work and the legal/policy domain. Finally, we will
conclude the paper in Section 4.
II. RELATED WORK
The work of Schlag discusses the use of unmanned aerial
vehicles by the government to conduct near constant
surveillance of individuals [4]. The author proposes a
consumer protection law that would designate authorized uses
of unmanned aerial vehicles by the government and
individuals. In his opinion, this law should help define the
public’s expectation of privacy from aerial surveillance by
drones. This work is a step in the right direction but is focused
on aerial vehicles only and on the protection of individual
privacy from the government, not privacy protection writ large.
Work is being done by a group at IBM to develop the
concept of a digital manifest for unmanned aerial vehicles [5].
The views of the author are his own and do not reflect the views of the
United States Military Academy, the United States Army, or the United
States Government.
This work proposes a technical solution to implement an
unmanned aerial vehicle manifest for onboard goods. The
manifest includes details about the origin and destination of the
vehicle and a listing of contents that could be displayed on the
vehicle or a remote terminal. The intent of this manifest
though is to assist an organization with the logistical challenges
of creating shipments, assigning transport, and then delivering
the goods. The manifest is not intended for use by law
enforcement.
The author of [6] imagines scenarios that may result in law
enforcement stops of an autonomous ground vehicle. The
scenarios include events such as passengers not wearing seat
belts, a vehicle that is reported stolen, and broken equipment.
These scenarios are the beginning of a more thorough
investigation into how unmanned systems may be used by
malevolent actors but it does not include scenarios such as
terrorist use of a vehicle. Furthermore, as highlighted in the
article, the authorities of law enforcement personnel to search
the vehicle are not fully defined and require the development of
policy and legal solutions.
In the article by Sullivan [7], the actions by states such as
California are highlighted. California is proposing laws that
require autonomous vehicles to have “law enforcement
interaction plans.” Developers of these systems would be
required to assist law enforcement by identifying vehicle
ownership and insurance. The open legal question is can a
manufacturer be forced to provide this information on a
privately owned vehicle? Again, these draft laws demonstrate
some initial thought on how autonomous vehicles operate in a
complex society, but they are not inclusive of all vehicle types
nor do they examine the most dangerous scenarios.
Kaminiski illustrates how many states are considering
privacy rules with respect to unmanned aerial vehicles [8]. The
range of state draft regulations on privacy in this area range
from treatment of surveillance in public areas as free speech to
more conservative approaches that require a subject’s consent
under various circumstances. A main point is that there is no
federal standard and only UAVs are considered in these
regulations. Additionally, most of these laws deal with
government surveillance drones and do not concern themselves
with citizen owned systems.
The MCity project attempts to foresee challenges with
regard to autonomous automobiles [9]. One of the products
they have released is a threat assessment tool to assist in
determining how vulnerable autonomous vehicles are to cyber-
attacks. Though the authors hypothesize about various
scenarios driverless vehicles could encounter, they only
propose possible technical solutions. Their work lacks non-
technical issue identification and only considers automobiles.
III. IDENTIFYING TECHNICAL, LEGAL, AND POLICY GAPS
It is clear that researchers are thinking about some of the
potential issues that unmanned vehicles will unearth.
However, most of the work deals with unmanned aerial
vehicles and how unmanned systems interact with the
government. Given that drones are now proliferate, this seems
reasonable. We propose that now is the time to begin research
and have discussions on all unmanned autonomous systems
and to extend that discussion beyond the government/public
interface. To assist in starting this discussion, we present the
following scenarios grouped into three main categories of
potential UV exploitation: the Normal UV, the Hacked UV,
and the Intrusive UV.
A. The Normal UV
The first category is UVs that are co-opted during the
completion of their normal tasks. The vehicle’s programmed
behavior and design attributes are used to commit criminal,
terrorist, or other illegal acts. This is illustrated in The Bomb
and The Criminal where well-behaved, normally operating
systems facilitate crime or acts of terrorism.
1) The Bomb
One evening, a food delivery robot makes a delivery to a
new customer. The robot makes its way through various
neighborhoods without difficulty passing several important
government facilities and even a few law enforcement
personnel. The now common presence of these robots and its
good behavior presents no cause for intervention in its
operation. At its destination, the food is accepted and payment
is made as normal. During the robot’s return trip, it suddenly
explodes outside of a local government office. Investigators
later conclude that the new customer deposited the explosive
inside the robot knowing that its return path would take it by
the desired target. Using a gps device to trigger the bomb and
the robot’s deterministically determined return path, the goal
of the terrorist was achieved.
Unmanned, autonomous delivery systems are currently
developed and in some cases in use [10]–[13]. The research
behind these systems includes the technical challenges of
navigation, collision avoidance, and the exchange of goods.
Items such as how users gain access to system compartments
have various solutions [10], [12], [14]. The challenges facing
autonomous systems are being solved and once conquered,
laws and regulations will most likely evolve to allow their use
in public. Before this happens, we need to explore the
scenarios for misuse to help design prevention mechanisms.
In this type of scenario, the unmanned system could have
many forms such as a taxi, a delivery vehicle, or a service
vehicle (e.g., a trash truck). The system in question could be
used passively like the delivery vehicle in the scenario. The
threat actors merely utilize the access granted to them and
publicly available knowledge of its operations to achieve their
goal. Alternatively, the system could be partly or completely
modified by the threat actor. In this situation, perhaps the
controller software is hacked to give control to the actor or the
system’s sensors are modified or spoofed to provide the
controlling system false sensor data.
Some of the research issues that arise from this scenario
include how to interrogate an unmanned system remotely to
determine its contents and mission. If the system is suspicious
or deemed a threat, how is the system disabled or stopped?
How is a criminal act attributed to the threat actor? The issues
are multi-faceted and include technical, legal, and policy
components. There is no single, easy solution.
Some technical solutions that assist in documenting system
contents and mission are found in the ideas in [5], [15], [16].
In the first solution, the legal marijuana industry is developing
digital manifests for law enforcement and auditing purposes.
The second solution is a proposed idea to create digital
manifests that will help companies’ logistical systems match
cargo to delivery systems. The last solution is an e-manifest
proposed by the EPA for hazardous waste cargos. This system
could allow for a check on potentially dangerous vehicles and
their routes. None of these solutions are sufficient to deal with
our scenario though. Manifests are not currently a requirement
for any unmanned systems and even those in the manned
systems world are usually only subject to inspection under
certain circumstances. Clearly more research work is needed
to develop solutions for these questions.
2) The Criminal
An autonomous taxi drives to a requested passenger pick-
up site. Upon arrival, it opens its doors to allow the passenger
to enter. Instead of a passenger boarding, a package is placed
into the vehicle. The vehicle then departs for its requested
destination. At that location, the doors open and the package
is removed by an unidentified individual. Unknown to the
proprietor of the taxi, the vehicle has just facilitated the
transportation of illegal substances. The account of the
requesting rider is spoofed and the identities of the highly
concealed originating and receiving agents is unknown.
In the present day, moving vehicles always have at least
one passenger - the operator. As such, law enforcement has the
ability to stop a vehicle, potentially search it, and if appropriate
hold the operator responsible for any violations of the law. In a
world of fully autonomous vehicles, this may no longer be true.
It is feasible that vehicles will not only be driverless, but also
without passengers. A vehicle could be used solely as a
delivery means for cargo. A UV could be sent to the dry
cleaners to pick-up an individual’s clothing and then return
with the cargo. In this scenario, who is responsible if the cargo
has illegal contents? Currently, many laws for vehicles are
based upon a driver being in control, and often require a
determination of intent [17]. How will this occur in a
passenger-less vehicle?
One of the first challenges is to attribute ownership of the
vehicle. This may be required to identify who is responsible
for the illegal action or if a vehicle is authorized to be in a
restricted location. A lot of effort is underway with regard to
identification of UAVs [18]–[21]. These efforts rely on
wireless transmitters and receivers that can interrogate drones
to determine ownership. So far, this effort has not been
extended to UGVs. Even with this technology, is it proper to
hold the owner responsible for the illegal act? What
technology needs to be emplaced to protect the owner from
these scenarios? If the owner is not responsible, how are all
interactions with the vehicle attributed in such a way that a
credible chain of responsibility is maintained? One suggestion
is that vehicle trips require "signatures" for programmed
route/destination instructions thus providing a log as to who is
responsible for the journey [17]. Without some form of
identifying technology, do police have the right to demand
vehicle data from the owner or manufacturer? Questions like
these are why some states, such as California, are requiring law
enforcement interaction plans when laws for autonomous
vehicles are being developed [7].
Another issue is what authorities does law enforcement
have to stop a suspicious vehicle? With a human driven
vehicle, pre-textual stops based on driving behavior such as
excessive speed, lane departure, or failure to signal provide an
avenue. These human errors will largely disappear if the
promised super-driving capabilities of autonomous vehicles is
delivered [22]. Some pre-textual stops may still be valid
though such as those based on broken equipment or expired
registration [6]. Reasonable suspicion is often enough to stop
a vehicle [23] as well. However, what now provides the basis
for that suspicion? The route a vehicle takes may simply be a
function of its navigation algorithm and may no longer indicate
potential misdeeds. Observation of the driver or passengers
when the vehicle is empty is no longer a basis for suspicion.
Some basis for stops will certainly remain, but the frequency of
such stops is sure to decrease [23].
If a vehicle is stopped, what authority does law
enforcement have to search it? How can consent to search be
obtained if no owner/operator is present (with or without
passengers)? Currently the driver is usually the individual who
can grant consent for a search. With the potential that there is
no driver, how will laws change to grant search consent?
Could the owner of the vehicle grant consent remotely? Could
states create laws that designate implied consent for search of
autonomous vehicles similar to searches that are done with
breath, blood, and saliva samples for suspected intoxicated
drivers? What will be considered probable cause for search
without consent? The “reason for arrest” often provides law
enforcement the mechanism to search for evidence [6]. Will
the records of a vehicle’s trips be searchable without warrant?
Autonomous vehicles are likely to store this information
locally or remotely so will that be considered “open
information” [22]? These are just a few of the legal and
regulatory issues that need to be examined prior to widespread
UV deployment.
B. The Hacked UV
The second category that we examine is UVs used
maliciously through hacking or other means of modification to
alter their normal operation. This is described in the
Bludgeon, the Plug, and the Kidnapper. These scenarios show
what is possible if the hardware, software, or sensing
capabilities of the UV are altered to allow the attacker to
control or change the behavior of the UV. Once the attacker
alters the UV, it can potentially participate in nefarious acts.
1) The Bludgeon
An autonomous ground vehicle drives up to a busy
intersection and stops at the red light. As pedestrians enter
the crosswalk in front of it, it suddenly accelerates into the
crowd striking multiple persons. It then proceeds inexplicably
to drive onto the nearby sidewalk striking more pedestrians
until it finally crashes into a building. The forensics team
concludes that the navigation system and key sensors have
been modified thus overriding the system’s normal safety
protocols. It appears that talented hackers have gained
access to the internals of the vehicle.
Any system that is operated through the use of computer
code and electronics is potentially vulnerable to this type of
scenario. As was seen with the Jeep hack of 2016 [24],
vehicles are no exception to this problem. Altering the
operations of autonomous vehicles from exploiting code to
simply modifying the sensors (ex. painting the camera lens) to
modifying the environment (ex. placing tape on portions of a
Stop sign to fool visual recognition) can take various levels of
effort and expertise. No matter the technique, a UV could be
altered to respond in this increasingly popular terrorist act.
Presently, an operator is the main mechanism for
preventing these types of occurrences. The operator is in a
position to prevent or stop modification of the vehicle while
they are present. When not present, they have the
responsibility to check the vehicle’s condition before
operation. So, without the operator, how do we know the
vehicle is not altered? How does the user/operator know that
the state of all embedded software is unaltered prior to
driving? What tools can be developed and emplaced that
allow civilians and law enforcement to visually or digitally
know the vehicle is operating within its designed
specifications? What requirements should be placed on
manufacturers to develop prevention mechanisms? What
liability should be attributed to the vehicle owner if it is
operated while altered? Should government create geofences
of areas and require vehicle manufacturers to enforce
compliance with these no-travel zones? The answer to all of
these questions deserve research and discussion before an
enterprising terrorist compromises a UV.
2) The Plug
While driving South out of Washington, D.C. early one
Friday evening, the traffic is worse than normal. Listening to
the traffic report on the radio, drivers are told of several
vehicles stopped in a row across all lanes of the road. No
drivers are reported in the vehicles, and no drivers were seen
exiting the parked cars. After the cars were removed from the
road and forensic investigation completed, it was discovered
that the vehicles were remotely piloted to the highway. Once
the vehicles were lined up side-by-side, the attackers stopped
the vehicles in the road, preventing traffic from moving.
Similar to hacking the vehicle in The Bludgeon scenario,
several vehicles could cooperate to disrupt traffic. This could
be a scenario where the vehicles drive slowly or are
completely stopped. Consider a section of road with concrete
barriers on one side, and a natural barrier on the other side
(mountain, cliff, water, etc.). California State Route 1, a
major highway along the Pacific Ocean, is an example of such
a road. If an attacker were able to commandeer some
unmanned vehicles, they would be able to prevent traffic flow
in either direction. This could potentially be done with a
single vehicle if parked across a two-lane road with no way to
drive around the vehicle. The Golden Gate Bridge in San
Francisco connects State Route 1 north and south of the
bridge. Using six commandeered vehicles would allow the
attacker to block all the lanes of the Golden Gate Bridge,
effectively disrupting up to 112,000 vehicles per day (source:
http://goldengatebridge.org/tolls_traffic/). New York City is
another example where an attacker could seriously impact
traffic patterns using a few cars to block one or more major
bridges or tunnels into the city.
All states have traffic laws requiring vehicles driving
slower than the normal speed to use the farthest right lane, and
several states now have laws requiring vehicles to move from
the far left, or passing, lane if a faster moving vehicle
approaches [25]. The laws are based on section 11-304 of the
Uniform Vehicle Code, a set of recommendations initially
developed for national standards for states to use [26]. How
will law enforcement officials engage with a vehicle violating
traffic laws, such as speeding or driving too slowly? A
Google self-driving car was pulled over for driving too slowly
in 2015 [27], although no ticket was issued by the officer. [6]
provides several examples where a UV might be pulled over,
but what if the vehicle does not respond to the flashing lights
of the police officer?
3) The Kidnapper
During a routine maintenance service for an unmanned
aerial system (UAS), a new “feature” is added that allows an
attacker to take control of the vehicle, in the same way that
researchers took control of ground vehicles in a 2015 and
2017 story in Wired magazine [24], [28]. During a routine
trip to work, the vehicle turns off course, carrying the
passenger to an isolated location. Communications on board
the vehicle, including voice and live video, are used to send a
ransom note to the victim’s family.
This scenario can be extended to ransom in other forms,
such as preventing operation of the vehicle, putting the
occupant in physical danger. Just as hospitals and home users
are encountering ransomware in their networked systems, UVs
can fall prey to this attack as well.
Several companies around the world are currently testing
working prototypes for UAVs which will function as air taxis
in cities, carrying passengers from roof top to roof top [29]–
[33]. One author discusses six different companies with
prototype unmanned aerial systems (UAS). Feist states “the
technology is ready, but the legal systems and safety for
passengers and people below have a long way to go yet.” [30]
He does not expect to see active air taxi systems in the United
States before 2020; however, Dubai, in the United Arab
Emirates, is pushing to have active systems before that, hoping
to be the first city in the world with autonomous air taxis.
Other experts consider a more reasonable timeframe for
ground vehicles to be 2030 [34], while more recent predictions
place the date as sometime in 2021 [35].
The technology to operate a UAS may be ready, but to
maintain safety and control of a high density of vehicles is not.
NASA is working on a system to handle a high density of
UAS, partnering with Uber to collect data. The system is the
UAS Traffic Management (UMT) network
(https://utm.arc.nasa.gov/index.shtml). [31]
If autonomous vehicles, whether air, ground, or sea
variant, can be compromised and remotely controlled by an
attacker, what can technology, policy and law do to prevent or
mitigate the damage? Can the occupant of the vehicle always
have manual override of the vehicle? Or will some form of a
“big red button” that puts the vehicle in a safe shutdown mode
be mandated?
C. The Intrusive UV
Lastly, we examine UVs that are operated by their owner to
perform legal, yet obtrusive and persistent surveillance of a
target, potentially violating individual privacy norms. The
Private Investigator and the Stalker are examples of this type of
UV exploitation. The ability of UVs to operate in public spaces
for extended durations may create an opportunity to infringe on
individual privacy that are not easily achieved in the present
day.
1) The Private Investigator
A private investigator is hired by a citizen to monitor the
actions of a selected individual. The investigator has
unmanned aerial and unmanned ground systems at her
disposal. Unlike in years past when the investigator’s
surveillance was limited to their personal time investment,
these new tools allow for constant surveillance of the target.
Ground systems monitor the target’s home and workplace
positioning themselves legally in public spaces. Aerial vehicles
monitor all of the target’s travels. There is no escaping the
unblinking eye of the unmanned vehicles and their cameras
and microphones.
Issues with regard to unmanned vehicles, mostly aerial,
have been well examined in the courts and this examination
will continue for the near future [36]–[39]. These cases
primarily deal with the issue of privacy with regard to the
government and the Fourth Amendment. The issue of the
legality of private citizens conducting surveillance on other
private citizens without their consent is not well explored.
When privacy is potentially violated by the government, the
salient issues revolve around factors such as the duration of the
surveillance, whether it is recorded, if the technology is
commercially available, whether the area of observation is in
the curtilage of the home, and if magnification/augmentation is
used. A popular framework used in some of these cases to
determine the legality of the surveillance is the “Mosaic
theory.” This suggests that a single person’s view of an
individual's movements over a period of time is limited and
therefore an expectation of movement privacy exists with
regard to that person [40]. Our scenario would fail the Mosaic
theory test. Clearly the targeted individual is under constant
surveillance and has no privacy of movement. Despite this, the
actions of the investigator would not be illegal because it’s not
the government conducting the surveillance.
The authors in [39] maintain that even if the government is
involved, the concept of "open fields", public areas and private
property that "do not provide the setting for those intimate
activities that the Fourth Amendment is intended to shelter
from government interference or surveillance", allow legal
surveillance. The private investigator in this scenario will
often be conducting their work in these open fields.
Additionally, in the case of US v Vela, it was ruled that using
night vision goggles to see into a vehicle on a public road was
allowable thus reinforcing the idea that surveillance in public
was o.k. even for the government [41]. Certainly if the
government is allowed to surveil individuals in public spaces,
then private citizens must be able to do it too. The question
remains though, is this something that we wish to condone?
Should the Mosaic theory apply to civilian surveillance? Can
citizens seek relief from the unwanted surveillance?
Some government agencies are providing a possible path in
this area. The United Kingdom's Regulation of Investigatory
Powers Act (RIPA) 2000 requires permission for a UAS to
record surveillance of a residential premise or private vehicle
and that surveillance system operations must notify individuals
of its operation so that copies can be requested if desired [42].
Indiana state law makes it a misdemeanor for anyone to place a
data recording device on another's property without consent (no
mention of public spaces) [43]. States such as Texas and
Missouri have considered laws prohibiting video of an
individual’s property without their consent [8]. Clearly there is
some concern for the privacy of individuals regardless of
government involvement. As such, coherent policies and
regulations need to be developed to cope with the surveillance
powers that unmanned vehicles provide citizens.
2) The Stalker
An individual becomes interested in the activities of another
person. He desires to not only monitor what that person does
and when, but also to disrupt their life. He directs unmanned
aerial vehicles to constantly fly nearby the target creating
noise pollution and recording all visible activity, some of which
is later posted on social media. Ground vehicles follow the
target and occasionally impede the free movement of the
person. Other ground vehicles position themselves outside the
person’s home and flood the wifi channel that the target’s
home router occupies thus limiting their ability to use the
internet. Finally, unmanned vehicles direct lights and lasers
into the windows of the target’s home at night making it
difficult to rest yet not violating any noise statutes.
This scenario shares some of the features of the private
investigator. The individual using the unmanned vehicles is not
a government official. The vehicles remain within public
spaces and thus are not trespassing. Deliberate attention is paid
to avoid violating local laws. The main difference here is the
intent of the stalker.
Technology is advancing quickly that provides features
enabling this scenario. A Google patent granted in 2015 [44]
describes a UAV system equipped with Automatic Target
Recognition and tracking. This technology could enable a
UAV owner to autonomously identify and follow their prey
without the interaction of a pilot. Additionally, the authors of
[45] describe an autonomous patrol system that is capable of
stalking ground-based personnel. Certainly the context of both
these systems is initially military in nature but as we have seen,
military technology frequently becomes civilian technology as
time passes.
Given current and future technology, what protections
should individuals have against unmanned vehicle
stalking/harassment? The limitation of the stalker’s time is no
longer a limiting factor on the harassment. A stalker can be
issued a restraining order but how would that apply to
unmanned vehicles? Adding to the difficulty of developing
policies and regulations to deal with this type of scenario is the
fact that anti-stalking laws are mostly a local affair.
Definitions of stalking vary, but terms such as willful,
malicious, repeated, unwanted are used in many. Intent, actual
harm done, and the perception of a “reasonable person” also
factor into these laws [46]. Overall, the lack of federal laws
regarding stalking create a challenging environment to regulate
unmanned vehicle use in these cases.
Despite this difficulty, it is clear that some action is being
taken. Cases in Arizona, Colorado, and Wisconsin have
resulted in individuals being punished for using technology
such as GPS trackers placed surreptitiously on vehicles to stalk
others [3]. As unmanned technology becomes more readily
available to the public, society needs to determine how it will
cope with its capabilities in these unforeseen scenarios.
IV. CONCLUSIONS
While experts argue about the exact timeframe for
autonomous systems to be commonplace, they do agree that
these systems are an eventuality. Technology is making rapid
progress, but our policies and laws have not kept pace. Will
manufacturers be allowed to place these systems in to
operation as soon as they are ready, or will the legal system
prevent them from operating until policy makers can decide
how to emplace safety measures and other controls in place?
Legal and policy regulations regarding liability may prevent
the implementation of unmanned vehicles [35].
Eventually, these systems will be commonplace in our
society. To avoid misuse of these UVs, it is time to research
and discuss solutions to potential problems such as those
mentioned in this work. We cannot afford to “develop and
then regulate” given the potential for harm. Scientists, policy
makers, and legal experts must unite to focus on prevention.
REFERENCES
[1] T. Ristenpart, N. Dell, K. Levy, and D. McCoy, “How domestic
abusers use smartphones to spy on their partners,” Vox, 21-May-2018.
[Online]. Available: https://www.vox.com/the-big-
idea/2018/5/21/17374434/intimate-partner-violence-spyware-
domestic-abusers-apple-google. [Accessed: 24-May-2018].
[2] US-CERT, “Heightened DDoS Threat Posed by Mirai and Other
Botnets,” 30-Nov-2016. [Online]. Available: https://www.us-
cert.gov/ncas/alerts/TA16-288A. [Accessed: 09-Dec-2016].
[3] J. Voelcker, “Stalked by satellite - an alarming rise in GPS-enabled
harassment,” IEEE Spectrum, vol. 43, no. 7, pp. 15–16, Jul. 2006.
[4] C. Schlag, “The New Privacy Battle: How the Expanding Use of
Drones Continues to Erode Our Concept of Privacy and Privacy
Rights,” Pitt. J. Tech. L. & Pol’y, vol. 13, p. 24, 2013.
[5] G. Hoareau, J. J. Liebenberg, J. G. Musial, and T. R. Whitman,
“Package transport container and transport operations for an
unmanned aerial vehicle,” US20160207627A1, 21-Jul-2016.
[6] J. F. Weaver, “Robot, Do You Know Why I Stopped You?,” Slate, 27-
Jun-2016.
[7] P. Bigelow, “On the Path to Autonomous Vehicles, Police Get Left
Behind | News | Car and Driver,” 27-Apr-2017. [Online]. Available:
https://www.caranddriver.com/news/on-the-path-to-autonomous-
vehicles-police-officers-get-left-behind. [Accessed: 30-Mar-2018].
[8] M. E. Kaminski, “Drone Federalism: Civilian Drones and the Things
They Carry,” Social Science Research Network, Rochester, NY,
SSRN Scholarly Paper ID 2257080, Apr. 2013.
[9] A. Weimerskirch and D. Dominic, “Assessing Risk: Identifying and
Analyzing Cybersecurity Threats to Automated Vehicles.” Mcity, Jan-
2018.
[10] “Marble, How We Do It,” Marble. [Online]. Available:
https://www.marble.io/. [Accessed: 30-Mar-2018].
[11] Panasonic Corporation, “Panasonic Autonomous Delivery Robots –
HOSPI – Aid Hospital Operations at Changi General Hospital in
Singapore,” 23-Jul-2015. [Online]. Available:
https://www.businesswire.com/news/home/20150723005636/en/Panas
onic-Autonomous-Delivery-Robots-%E2%80%93-HOSPI-
%E2%80%93. [Accessed: 30-Mar-2018].
[12] “Starship.” [Online]. Available: https://www.starship.xyz/. [Accessed:
30-Mar-2018].
[13] “Why the wait for delivery drones may be longer than expected,” The
Economist, vol. 423, no. 9044, pp. 10–13, 10-Jun-2017.
[14] “How it works — Dispatch.” [Online]. Available:
http://dispatch.ai/how-it-works/. [Accessed: 30-Mar-2018].
[15] O. US EPA, “Learn About the Hazardous Waste Electronic Manifest
System (e-Manifest),” US EPA, 05-Mar-2016. [Online]. Available:
https://www.epa.gov/e-manifest/learn-about-hazardous-waste-
electronic-manifest-system-e-manifest. [Accessed: 30-Mar-2018].
[16] “Metrc | The System,” Metrc, The System. [Online]. Available:
https://www.metrc.com/the-system. [Accessed: 11-Jun-2018].
[17] F. Douma and S. A. Palodichuk, “Criminal Liability Issues Created by
Autonomous Vehicles,” Santa Clara Law Review, vol. 52, no. 4, p.
1157, Dec. 2012.
[18] M. McFarland, “White House pilot program to allow more drone tests
- Oct. 25, 2017,” 25-Oct-2017. [Online]. Available:
http://money.cnn.com/2017/10/25/technology/business/drones-trump-
local/index.html. [Accessed: 30-Mar-2018].
[19] DJI, “‘What’s In a Name?’A Call for a Balanced Remote
Identification Approach,” Dropbox, 22-Mar-2017. [Online].
Available:
https://www.dropbox.com/s/v4lkyr2kdp8ukvx/DJI%20Remote%20Ide
ntification%20Whitepaper%203-22-17.pdf?dl=0. [Accessed: 30-Mar-
2018].
[20] C. Ramsey, “AIR6388 (WIP) Remote Identification and Interrogation
of Unmanned Aerial Systems - SAE International.” [Online].
Available: https://www.sae.org/standards/content/air6388/. [Accessed:
30-Mar-2018].
[21] S. French, “Drone identification: What we know about the FAA ARC
plans so far,” The Drone Girl, 30-Jun-2017. [Online]. Available:
http://thedronegirl.com/2017/06/30/drone-identification-faa-arc/.
[Accessed: 30-Mar-2018].
[22] O. Kerr, “Opinion | How self-driving cars could determine the future
of policing,” Washington Post, 16-Jun-2017.
[23] R. Roseman, “When Autonomous Vehicles Take over the Road:
Rethinking the Expansion of the Fourth Amendment in a Technology-
Driven World – Richmond Journal of Law and Technology.”
[Online]. Available: http://jolt.richmond.edu/2014/01/06/when-
autonomous-vehicles-take-over-the-road-rethinking-the-expansion-of-
the-fourth-amendment-in-a-technology-driven-world/. [Accessed: 25-
Apr-2018].
[24] A. Greenberg, “A Deep Flaw in Your Car Lets Hackers Shut Down
Safety Features | WIRED,” 16-Aug-2017. [Online]. Available:
https://www.wired.com/story/car-hack-shut-down-safety-features/.
[Accessed: 23-Apr-2018].
[25] A. Essex, D. Shinkle, and A. Teigen, “Transportation Review |
Speeding and Speed Limits,” 16-Feb-2017. [Online]. Available:
http://www.ncsl.org/research/transportation/transportation-review-
speed-limits.aspx#Keep%20right. [Accessed: 21-May-2018].
[26] “Uniform Vehicle Code,” National Committee on Uniform Traffic
Control Devices. [Online]. Available:
http://www.ncutcd.org/Pages/default.aspx. [Accessed: 21-May-2018].
[27] D. Melvin, “Cop pulls over Google self-driving car, finds no driver to
ticket,” CNN, 13-Nov-2015. [Online]. Available:
https://www.cnn.com/2015/11/13/us/google-self-driving-car-pulled-
over/index.html. [Accessed: 28-May-2018].
[28] A. Greenberg, “Hackers Remotely Kill a Jeep on the Highway—With
Me in It,” WIRED, 21-Jul-2015. [Online]. Available:
https://www.wired.com/2015/07/hackers-remotely-kill-jeep-highway/.
[Accessed: 23-Apr-2018].
[29] M. Moon, “Dubai tests a passenger drone for its flying taxi service,”
Engadget, 26-Sep-2017. [Online]. Available:
https://www.engadget.com/2017/09/26/dubai-volocopter-passenger-
drone-test/. [Accessed: 07-May-2018].
[30] J. Feist, “Drone taxi service - passenger drones - DroneRush,” 13-
Apr-2018. [Online]. Available: https://www.dronerush.com/drone-
taxi-passenger-drones-10666/. [Accessed: 30-May-2018].
[31] L. Greenemeier, “Here’s What’s Needed for Self-Flying Taxis and
Delivery Drones to Really Take Off,” Scientific American, 15-May-
2018. [Online]. Available:
https://www.scientificamerican.com/article/heres-whats-needed-for-
self-flying-taxis-and-delivery-drones-to-really-take-off/. [Accessed:
13-Jun-2018].
[32] M. Margaritoff, “Watch the Ehang 184 Passenger Drone Successful
Taxi Someone Around,” The Drive, 05-Feb-2018. [Online]. Available:
http://www.thedrive.com/aerial/18261/watch-the-ehang-184-
passenger-drone-successfully-taxi-someone-around. [Accessed: 30-
May-2018].
[33] A. J. Hawkins, “Airbus’ autonomous ‘air taxi’ Vahana completes its
first test flight,” The Verge, 01-Feb-2018. [Online]. Available:
https://www.theverge.com/2018/2/1/16961688/airbus-vahana-evtol-
first-test-flight. [Accessed: 11-Jun-2018].
[34] S. Underwood, “Automated, Connected, and Electric Vehicle
Systems,” p. 154, Dec. 2014.
[35] J. Walker, “The Self-Driving Car Timeline - Predictions from the Top
11 Global Automakers,” TechEmergence, 29-May-2017. [Online].
Available: https://www.techemergence.com/self-driving-car-timeline-
themselves-top-11-automakers/. [Accessed: 13-Jun-2018].
[36] A. Cavoukian, Privacy and drones: Unmanned aerial vehicles.
Information and Privacy Commissioner of Ontario, Canada Ontario,
Canada, 2012.
[37] Marc Jonathan Blitz, “The Fourth Amendment Future of Public
Surveillance: Remote Recording and Other Searches in Public Space,”
Am. U. L. Rev., vol. 63, no. 1, 2013.
[38] T. Dunlap, “We’ve Got Our Eyes on You: When Surveillance by
Unmanned Aircraft Systems Constitutes a Fourth Amendment
Search,” S. Tex. L. Rev., vol. 51, pp. 173–204, 2009.
[39] J. J. Vacek, “Big Brother Will Soon Be Watching—or Will He?
Constitutional, Regulatory, and Operational Issues Surrounding the
Use of Unmanned Aerial Vehicles in Law Enforcement,” N.D. L.
Rev., vol. 85, p. 20, 2009.
[40] A. B. Talai, “Drones and Jones: The Fourth Amendment and Police
Discretion in the Digital Age,” Cal. L. Rev., vol. 102, p. 729, 2014.
[41] B. Stubbs, “Technological Ubiquity and the Evolution of Fourth
Amendment Rights,” Drake L. Rev., vol. 62, pp. 575–598, 2014.
[42] R. L. Finn and D. Wright, “Unmanned aircraft systems: Surveillance,
ethics and privacy in civil applications,” Computer Law & Security
Review, vol. 28, no. 2, pp. 184–194, Apr. 2012.
[43] A. Bohm, “Status of 2014 Domestic Drone Legislation in the States,”
American Civil Liberties Union, 30-Jun-2014. [Online]. Available:
https://www.aclu.org/blog/status-2014-domestic-drone-legislation-
states. [Accessed: 30-Mar-2018].
[44] K. Kokkeby, R. Lutter, M. Munoz, F. Cathey, D. Hilliard, and T.
Olson, “Methods for autonomous tracking and surveillance,” 12-Dec-
2008.
[45] J. Oyekan and H. Hu, Towards Autonomous Patrol Behaviours for
UAVs. .
[46] E. Petch, “Anti-stalking laws and the Protection from Harassment Act
1997,” The Journal of Forensic Psychiatry, vol. 13, no. 1, pp. 19–34,
Jan. 2002.
