papers reflection

1 Self-driving cars: A case study in making new markets

December 2012

Self-driving cars

A case study in making new markets

Luis Araujo, Katy Mason and Martin Spring

Lancaster University

2 Self-driving cars: A case study in making new markets

This essay considers the prospects for the future of the self-driving car, and looks in

particular at the issues around creating a functioning market for such vehicles. The aim

is to shed light on the principles and practicalities of ‘market making’, a discipline in

which the authors are experts. This essay is released alongside, and intended as an

accompaniment to, a longer paper by the Big Innovation Centre entitled: ‘Market

Making: A modern approach to industrial policy’. This report can be accessed at

http://www.biginnovationcentre.com/Publications.

The essay was written for the Big Innovation Centre by the following authors, all

academics at Lancaster University:

Professor Luis Araujo, Personal Chair in Marketing, Lancaster University

Dr. Katy Mason, Reader in Marketing, Lancaster University

Dr. Martin Spring, Senior Lecturer in Operations Management, Lancaster

University

The Big Innovation Centre is an initiative of The Work Foundation and Lancaster

University. Launched in September 2011, it brings together a range of companies,

trusts, universities and public bodies to research and propose practical reforms with the

ambition of making the UK a global open innovation hub as part of the urgent task of

rebalancing and growing the UK economy, and with the vision of building a world-class

innovation and investment ecosystem by 2025.

For further details, please visit www.biginnovationcentre.com

3 Self-driving cars: A case study in making new markets

Self-driving cars: coming to a road near you?

The recent and much-publicised bills to authorise driverless cars in the US states of

Nevada and California has raised a number of interesting questions about the nature of

this innovation and the potential market for these vehicles 1 . Would self-driving cars

make a big difference to our daily lives? Is it realistic to expect a market for them to

emerge?

The two questions are interrelated. We can glean from our knowledge of innovation that

the less disruptive innovations are, the more they can make use of or insert themselves

seamlessly into the existing socio-technical infrastructures, the greater their chances of

success 2 . To go against the grain and cause serious disruption often requires major acts

of faith and the development of large technical systems, to borrow Thomas P. Hughes’

term 3 . In short, the less disruptive a technology is deemed to be, the quicker and wider

the market for it can grow, ceteris paribus.

How disruptive are driverless cars? To adopt a purely technological angle, the short

answer is not much. At one level, these events can be seen in the context of the

inexorable march of machines to acquire skills we have hitherto regarded as inherently

human 4 . The notion of driverless vehicles in other modes of transport has been around

for a while. An early example is the Parisian Aramis project. Aramis was an advanced

form of personal transport developed for RATP (Régie Autonome des Transports

Parisiens) in Paris, between 1969 and 1987, by the French defence company Matra.

Aramis was designed to combine the most attractive aspects of personal and public

transport, with small, driverless carriages combining to form a flexible train whilst in

transit. The vision behind Aramis was that users could summon carriages from passing

trains which would then rejoin the train via contactless linkages. The flexible nature of

the Aramis train would remove unnecessary stops and transfers, combining the best

aspects of conventional cars with the more ecologically-sound benefits of public

transport 5 .

Whereas Aramis required a dedicated infrastructure (eg stations, rails) and proved to be

a somewhat expensive failure, driverless cars can more appropriately be seen as an

evolution of an existing artefact – the conventional passenger car – even if they partly

share Aramis’ ideal. In the last few decades, cars have literally been swamped with all

forms of electronic systems. From engine management systems, to trip computers, and

1 http://www.nytimes.com/2012/01/24/technology/googles-autonomous-vehicles-draw-skepticism-at-legal-

symposium.html; http://www.bbc.co.uk/news/technology-19726951; http://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=201120120SB1298&search_keywords= 2 See Urry, J. (2004). The ‘System’ of Automobility. Theory, Culture & Society 21(4-5): 25-39.

3 Hughes, T. P. (1983). Networks of Power: Electrification in Western Society. Baltimore, MD., John Hopkins

University Press. 4 http://www.guardian.co.uk/technology/2012/sep/30/google-self-driving-car-unemployment?INTCMP=SRCH

5 Latour, B. (1996). Aramis, or the Love of Technology. Cambridge, MA., Harvard University Press.

4 Self-driving cars: A case study in making new markets

self-diagnostic tools, passenger cars have evolved to the point where they can be seen

as a hybrid, a complex assemblage of mechanical and electronic systems. In the more

recent past, we have seen automakers embedding technologies such as Bluetooth

connections, GPS navigation systems, in-vehicle infotainment and wi-fi systems and so

on 6 .

The leap to the next stage – driverless cars – may not be as revolutionary or as

potentially disruptive when seen from this long-term, evolutionary perspective. The

European Commission project known as SARTRE – Safe Road Trains for the

Environment – has already demonstrated that driverless vehicle trains in motorways

seem to work safely. Cars in these trains are fitted with features such as cameras, radar

and laser sensors – allowing a ‘carriage’ vehicle to monitor the ‘locomotive’ and other

vehicles in their immediate vicinity. The most interesting aspect of this project is to show

that there is little or no change required in either the vehicles or the infrastructure for the

train to work. All that is required is a wireless network between cars in the train and the

appropriate software to ensure that, for example, gaps between vehicles travelling at

speed (~52 mph) remain at around 20 feet 7 .

Google has recently taken these ideas to yet another level. The Google driverless

system combines information gathered from Google Street View with artificial

intelligence software, combining inputs from video cameras inside the car, a LIDAR

(light detection and ranging) sensor positioned (prominently) on top of the vehicle, radar

sensors on the front, and a position sensor attached to one of the rear wheels to help

locate the car's position. It has been in gestation for a while – benefiting from Google’s

close links with Stanford University – and is effectively designed to replicate the

capabilities of human drivers – eg in terms of seeing and sensing road conditions and

other road users – while mitigating or eliminating altogether the problems associated

with human error and miscalculation.

In summary, although the Google driverless car can be regarded as a disruptive

technology in the sense that it ultimately does away with one of the linchpins of the

automobility system (ie the licensed, legally responsible driver linking the car to the road

and traffic management infrastructure), it is very much the product of an incremental

evolution and convergence of a number of different technologies. The more interesting

questions are:

• How does this change impact upon the broader automobility system, of which

the car-driver combination is a key component?

• What is the potential market for driverless cars?

6 http://www.independent.co.uk/life-style/motoring/wifi-cars-hitting-the-information-superhighway-2260187.html

7 http://news.bbc.co.uk/1/hi/8349923.stm

5 Self-driving cars: A case study in making new markets

• Will driverless cars become as common as driven cars are now, with driverless

progressively displacing driven? Will driverless cars meet the same fate as

Aramis? Could they instead be confined to a niche (e.g. catering for those with

mobility problems or disabilities, taxi ranks in smart cities)? Or, will we have

driverless / driven hybrids with driverless features switched on and off (eg during

long motorway journeys or traffic jams) as circumstances dictate?

The potential benefits of self-driving cars

Before we tackle these questions, let us consider the arguments for driverless over

driven cars. There are three main selling points for driverless cars:

1) Fewer accidents. A significant proportion of accidents and loss of life on the

roads are due to driver error. Driverless cars could drastically reduce the

accident toll. Lives saved is the measure that will conclusively show the

advantages of driverless over driven cars.

2) More productive commutes. A significant percentage of the population in

developed, car-rich economies spend considerable time commuting to work.

Driverless cars would enable these unproductive hours and minutes to be

converted into productive work and/or leisure time. GDP and wellbeing

indices should go up accordingly.

3) Fewer traffic jams. Driverless cars would be better adapted to higher

volumes of traffic, as they would be able to travel at higher speeds while

keeping shorter distances between vehicles. Decreased congestion and

better overall fuel economies will be achieved as a result 8 .

Moreover, driverless cars could extend ownership to new groups of people (eg the

visually impaired) and protect road users from well-documented hazards (eg drink

drivers) 9 . Eventually, insurance premiums could drop to negligible amounts even for

neophyte drivers, assuming we will retain the practice of training people with skills that

they may rarely need to call upon.

Barriers to the growth of a market for self-driving cars

But there are plenty of uncertainties about this transition to a driverless future. The

current socio-technical infrastructure around cars is closely associated with a complex

system of practices that may change in unpredictable ways. Will driverless cars reduce

congestion on the roads? Possibly yes, but maybe not. Driverless cars could lead to an

8 http://www.bbc.co.uk/news/magazine-18012812.

9 As Sergey Brin, one of Google’s founders, put it: “"This has the power to change lives. Too many people are

underserved by the current transport system. They are blind, or too young to drive, or too old, or intoxicated." Quoted in: http://www.theregister.co.uk/2012/09/25/google_automatic_cars_legal/

6 Self-driving cars: A case study in making new markets

increase in short journeys into crowded city areas bypassing parking charges (eg send

the driverless car round the block while you wait for your take-away meal or pick up the

dry cleaning). City centres could be filled with driverless, empty cars negotiating traffic

jams without a passenger in sight.

Maintenance networks

The existing socio-technical infrastructure around the car was built slowly, the range of

distances people were prepared to travel expanding as more and better roads were

built, and as filling stations and maintenance networks were gradually put in place. We

easily forget how much our everyday use of the car is wholly dependent on this

infrastructure. Kevin Borg starts his book Auto Mechanics with a suggestive scene

setting paragraph: “Cars break down. They always have. On a warm spring day in 1901,

a man named Robin Damon expected to enjoy the new freedom of automobility – swift

individual travel without rails, without schedule, and free of wilful horses. Instead, he and

his friends spent six hours in the hot sun replacing spark plug gaskets, putting in new

ignition points, and replacing a broken battery wire in the friend’s stranded ‘gasoline

carriage’. The promises of the new technology, it turned out, were conditional” 10

.

This vignette reminds us that cars – driven or driverless – still need be maintained and

repaired. If anything, driverless cars with their intricate electronic systems and complex

interaction between electronic and mechanical systems might be prone to difficult-to-

diagnose and difficult-to-cure problems. The ability of rescue patrols to repair cars on

the spot will probably be quite limited and the skills and equipment of service

dealerships will need to be significantly upgraded. Driverless could become a synonym

for undriveable.

Cyber-security risks

Even if the electronics in driverless cars prove to be reliable – or at least, as reliable as

contemporary mechanical systems – there are other dangers lurking. A joint report by

Internet security firm McAfee and Wind River calls attention to the security risks involved

in embedded car devices 11

. More specifically, the report argues that experiments such

as the Google driverless car and smart roads (eg equipped with traffic or speed

sensors) demonstrate the potential of coordinated, connected communication from all

sorts of electronic systems within the car. However, little has been done to ensure the

security of these systems. It is worth reminding ourselves that the first remote keyless

entry systems did not implement any security and were easily compromised: a regular

learning universal remote control for consumer electronics was able to record the key

signal and replay it at a later time. This was only corrected after some painful lessons

10

Borg, K. L. (2007). Auto Mechanics. Technology and Expertise in Twentieth Century America. Baltimore, MD., John Hopkins University Press, p. 1. 11

http://www.mcafee.com/us/resources/reports/rp-caution-malware-ahead.pdf

7 Self-driving cars: A case study in making new markets

and examples of high-tech security loopholes persist to this day 12

. The increasing use of

embedded and connected devices in cars could lead to serious problems from remote

attacks via Bluetooth to the hacking of private data stored in the car’s infotainment

system.

Legal liability

Lastly, there is the issue of liability in the case of accidents. The law is notoriously slow

in catching up with new technologies and the novel practices they enable (eg defining

and enforcing copyright in a digital age). There are a number of potentially hazy areas

such as how the hardware and software involved interact with the car’s mechanical

systems; product liability issues with different manufacturers who take crucial roles in

putting the system together (eg is it a Google car or a Toyota Prius driverless-enabled

by Google?); the role of other cars and road users such as pedestrians or cyclists; and

the responsibility of the owner to perform proper maintenance on the vehicle and its

automated systems.

What would happen in the case of a collision say between a driven and a driverless car,

or even between two driverless cars? What would insurers be prepared to cover and

what types of arguments would the law be prepared to hear in cases of damages and

personal injuries? And what exactly would insurers insure? Drivers will still matter, at

least as owners of vehicles with ultimate responsibility for their maintenance, with the

possibility that manual overriding will take precedence over driverless features as far as

the law is concerned 13

.

From self-driving cars to other disruptive technologies: lessons for market

makers

The market for driverless cars does not just depend on manufacturers persuading

legislatures to pass bills and a reluctant public to give up driving. For driverless cars to

realise their commercial potential, they have to fit with the current automobility system or

the system needs to be reconfigured to accommodate a novel entity. Some of those

adaptations will require minor changes in the practices that make use of cars in our

everyday lives, whilst others will need major investments in changing or upgrading

existing networks (eg service and maintenance).

This example illustrates how markets are enmeshed in a complex matrix of socio-

material infrastructures and gradually shaped by a multiplicity of actors with different

12

http://www.channel4.com/news/bmw-security-loophole-leads-to-rise-in-car-thefts 13

California’s Senate Bill 1298 (Vehicles: autonomous vehicles: safety and performance requirements) states: “The driver shall be seated in the driver’s seat, monitoring the safe operation of the autonomous vehicle, and capable of taking over immediate manual control of the autonomous vehicle in the event of an autonomous technology failure or other emergency”.

8 Self-driving cars: A case study in making new markets

agendas and priorities. Markets provide efficient solutions for problems that can be

framed and disentangled from all sorts of externalities and controversies (eg protests by

pressure groups, spokespersons for future generations). But their strength is often their

limit. Markets can easily marginalise, forget or ill-treat those whose concerns have been

excluded for one reason or another and produce all forms of externalities 14

. The

example of driverless cars provides another instance of how innovations affect a variety

of constituencies whose interests and concerns may be challenged by the innovation 15

.

Ultimately, driverless cars may yet turn out to be the right answer to the wrong question

if the contemporary challenge is how to get us out of energy-intensive modes of

personal transport to more efficient and ecologically-sound modes of public transport.

The story of Aramis, with its visionary compromise between private and public modes of

transport, may yet have a happy ending. The British company Ultra has designed a

personal rapid transit (PRT) system, based on driverless pods capable of carrying 4

passengers and their luggage, linking Heathrow’s T5 with the business car park. The

pods are battery-powered, driverless vehicles offering a novel way to travel to and from

the terminal 16

. Could this yet be the future of driverless vehicles?

14

Callon, M., P. Lascoumes, and Y. Barthe (2009). Acting in an Uncertain World. Cambridge MA, The MIT Press. 15

Somewhat prematurely perhaps, driverless cars have already featured in political campaigning: http://www.youtube.com/watch?feature=player_embedded&v=NUuBXCEWOhc 16

http://www.ultraglobalprt.com/

9 Self-driving cars: A case study in making new markets

Contact details

Big Innovation CenBig Innovation CenBig Innovation CenBig Innovation Centretretretre

The Work Foundation

21 Palmer Street

London SW1H 0AD

[email protected]

www.biginnovationcentre.com

www.theworkfoundation.com

All rights reserved © Big Innovation Centre (The Work Foundation and Lancaster University). No part of this

publication may be reproduced, stored in a retrieval system or transmitted, in any form without prior written

permission of the publishers. For further information please contact [email protected]. The Work

Foundation Alliance Limited, 21 Palmer Street, London, SW1H 0AD, UK. Registered Charity No. 1146813.

Registered as a company limited by guarantee No. 7746776.
Registered address: Lancaster University, Bailrigg, Lancaster LA1 4YW, UK.

10 Self-driving cars: A case study in making new markets