The development of driverless vehicles is fast, and the technology has the potential to significantly affect the transport system, society and environment

1. Autonomous Vehicle Implementation
Abstract: The development of driverless vehicles is fast, and the technology has the potential to
significantly affect the transport system, society and environment. However, there are still many open
questions regarding what this development will look like and there are several counteracting forces. This
paper addresses the effects of driverless vehicles by performing a literature review of twenty papers that
use simulation to model effects of driverless vehicles. By combing and analyzing the results from these
simulation studies, an overall picture of the effects of driverless vehicles is presented.
The reviewed papers providea first prediction of factors such as waiting time, VKT andtrip cost, for urban
areas and for schemes where there is one service providerpresent. To get a deeper understandingof the
effects of driverless vehicles, aspects suchaslocal spatialconsiderations, e.g. atpick-upstations, andmore
complex schemes with competitionbetween service providersshouldbe studied. Furthermore, there is a
need for sensitivity analyses regarding travel demand.
Keywords:Driverlessvehicle;Automatedvehicle;Autonomoustaxi;Traffic simulation; Societal effects;
Introduction
The developmentof automateddrivingtechnology andits use in driverless automatedvehiclesis fast, and
the technology has the potential to significantly affect the transport system, society and environment.
However, there are still many open questions regarding what this development will look like and several
counteracting forces exist (Pernestål Brenden, Kristoffersson, and Mattsson 2017; Milakis et al. 2017;
Townsend2014). Forexample, automationmay leadtoincreasedroadcapacity, whichreducescongestion.
On the other hand, the possibility to use the time in the car for other things than driving, lower marginal
travel costs and new user groups may lead to increased traffic (Litman 2015). The societal effects of
driverless vehicles do not come directly from the technology itself, but rather from how it is used (Barth,
Boriboonsomsin, andWu2014;Brown, Gonder, andRepac2014;MacKenzie, Wadud, andLeiby 2014)and
there is still a lack of understanding how the contradicting forces interact.
In the literature there are several simulations studying different aspects of effects of driverless vehicles.
Each of thesesimulationstudiesbea case study. By reviewingthe simulationstudies, comparingthem, and
acknowledging that they describe different cases, the aim of this paper is to achieve a holistic picture of
the effects of driverless vehicles. By using this approach, the paper addresses questions such as: Which
type of application areas and mobility concepts are covered by the existing simulation studies? What are
theeffects onperformanceindicatorssuchastrip cost, vehiclekilometers travelled (VKT), fleet size, waiting
time etc.? What are the existing research gaps?
Methodology
This literature review uses Wee and Banister (2016) as methodological framework. Database searches in
combination with forward and backward snowballing (Jalali and Wohlin 2012) have been used as search
strategies to find relevant papers. The search was performed with the keywords“autonomousvehicle(s)”

2. and “driverless vehicle(s)” combined with “impact AND service”, “taxi”, “fleet size” and “model” in title,
abstractandkeywordswere usedas search termsin Scopus andGoogleScholar. This initial search resulted
in a set of fifteen papers consideredrelevant for the scopeof thisliterature review. With thesepapers asa
base, anotherfive relevant paperswere foundvia backward andforward snowballing. Thus, the literature
search resulted in twenty papers reviewed in this article.
To make an overview of the reviewed papers, nine dimensionsare selected: simulationapproach, scale of
application, mobility concept, penetrationrate, travel demand, trip cost, vehicle kilometers travelled, fleet
size, and waiting time (see Table 2). The first five dimensions are chosen to compare the set-up of the
simulation studies. The other dimensions are chosen to compare reported effects of driverless vehicles.
The resultdimensionsarechosenby identifyingwhich are the mainvariablesfor which resultsare reported
in the reviewed papers.
CONTEMPORARY PROGRESS:
The modern automobile companies keep coming up with newer autonomous features in their recent
models. Technologicaladvancementsseenevery day in areas like informationtechnology, communication,
data analysis and storage etc. is not exclusive to these areas alone. The realm of autonomouscars is also
progressingata rapid rate these days. Segway IncorporatedandGeneralMotorsjointly developeda 2 seat
electric car, basically designed for urban environments and which could be driven normally or operated
autonomously. Known as GM’s EN-V (General Motor’s Electric Networked Vehicle), it was first unveiled
from 1stMay through31stOctober2010atthejoint GM & SAICpavilionat the Expo 2010 inShanghai. EN-
V was further divided into three different vehicle types:Jiao (Pride), Miao (Magic), andXiao (Laugh). EN-V
exhibitsautonomousfeaturessuchasself- parking/retrieval, vehicleplatoonsandcollisionavoidance. GM’s
EN-V became an important advancement towards paving the way in realizing a higher grade of vehicle
connectivity, vehicle interfaces, motion control algorithms, and connected autonomous driving
architecture (Eberle, 2011; Mudalige, 2010.
The first cars licensed for autonomous driving on the streets and highways of the German state of Berlin
are Made In Germany and Spirit of Berlin, developed by the Auto NOMOs Labs. It was a project of Freie
Universität, Berlin, andfundedby theGerman Federal Ministry of EducationandResearch. The project had
developing technology for driver assistance systems, innovative safety systems for cars and full
autonomousvehiclesin airportsor mines as its major objectives. Ithas a very accurate GPS unit andthree
laser scannersat the front, andthree at the rear of the vehicle detect any car or pedestrian all aroundthe
car. It can also detect traffic lights, intercity traffic and roundabouts (Reuschenbach, 2011; Dias, 2013).
Navia is a robotically driven electric shuttle which operates at a maximum speed of 20 kilometers per
hour. Made by Induct Technology, France, it can accommodate 10 passengers. It uses four LIDAR
units andstereoscopic optical cameras, andit does not require any road modifications. Its LIDAR unit and
optical cameras help in generating a real-time three-dimensional map of the surroundings. It is being
successfully testedat variousuniversitiesacross Switzerland, Englandand Singapore(Zhang, 2014). Google
plans to unveil hundred driverless car prototypes built inside Google's secret X lab, as
officially said by the company on 27th May, 2014. Google proclaims it to be a manifestation of years of
work that beganby modifyingexisting vehicles. Google plansto release thesemodelsin the yearsto come.
The field of autonomousautomationkeeponexpandingday by daywithneweradvancesfora futurewhich
is safer and more efficient (Headrick,2014; Sunwoo, 2014).

3. CONCLUSIONS
This paper discusses basic chronology leading to the development of autonomous cars. Autonomous
vehicles developed from the basic robotic cars to much efficient and practical vision guided vehicles.
The developmentofMercedes- Benzvisionguidedautonomousvanby ErnstDickmannsandhisteamgave
a paradigm shift to the approach followed in autonomous cars. Also, contemporary developments in
autonomous cars reflect the vivid future autonomous cars behold. Official future predictions about
autonomous cars point out that most automobile companies will launch cars with semi and fully
autonomous featuresby 2020. Mostcarsareexpectedtobefully autonomousby2035,accordingtoofficial
predictions as cited earlier. This paper reviewed the historical antecedents, contemporary advancements
and developments, and predictable future of semi and fully autonomous cars for public
use.