IEEE Robotics & Automation Magazine - March 2011 - 66

At the implementation stage, usually the individual
employs the innovation to a varying degree that depends
on the situation and is determined by the usefulness of
the innovation and may possibly search for further information about it. In our case, the local municipality organized a second public meeting; this time, the meeting
included training on how
* to use the DustCart robot
and request its service and
During the testing of
information on its usefulness and ease of use.
DustCart, no accidents
Finally, as to confirmation, the citizens finaloccurred, and the robot
ized their decision to use
DustCart for the entire
proved to be reliable
period of the test, using it
to its full potential.
and safe.
By investigating the
* impact
of political factors upon urban policy outputs, the relationship between
government and citizens' decisions has been found to be
very important in the adoption of the robot and its service.
In this process, the organizational environment and the
local government were significant variables and directly
influenced the testing of the innovation DustCart, where
many other predictors of the use of this innovation would
have been expected to be more important. These factors
cannot explain or predict the success or failure of the
future use of DustCart, but they certainly played a fundamental role in this test.

During the testing of DustCart, no accidents occurred, and
the robot proved to be reliable and safe.
In this article, we have attempted to highlight that there
also exist nontechnological challenges for deploying service robots in urban areas. In particular, we have pointed
out the implications related to legal regulations and social
acceptance.
As far as the legal issue is concerned, as we have pointed
out, the legal classification of autonomous robots at the level
of road traffic code still remains a problem. The solutions
that were adopted in Peccioli are valid only for a temporary
test. With regard to social acceptance, we believe that DustCart was accepted by the inhabitants of Peccioli, and this
was on account of the fact that people associated its presence
with the accomplishment of an important service: separate
waste collection (Figure 9). Citizens also appreciated the fact
that DustCart offered an on-demand service; in other words,
it fitted the needs of each person and provided a door-todoor service, which meant that people did not have to move
away from their homes or shops to dispose of rubbish. This
last feature was quite important, especially, for elderly people
and for those working in shops.
However, although not a demonstration, but a real
deployment, the testing of DustCart was limited in time
and took place under partially controlled conditions. In
other words, there was neither sufficient time nor the
appropriate conditions to find out other potential ethical,
legal, and social implications. For instance, it was not possible to find out the existence of abuses or improper behaviors toward the robot, such as vandalism. It has been
discussed earlier [23] that vandalism can affect the safety
as well as quality of service provided by robots operating in
urban environments. A very simple example of robot vandalism is a group of people surrounding the robot and blocking
its way. In addition, it would have been interesting to evaluate dustmen's acceptance level in case the robot was used
permanently or that of road users. As discussed earlier, one
of the main problems encountered during the testing of the
robot was traffic congestion. Despite the presence of a robot
lane, the streets of Peccioli are too small for both robots and
cars. In conclusion, among the lessons learned from Peccioli
that are useful for paving the way for robot deployment in
urban settings is that acceptance is no more a matter of the
users willingness to use a product [24]. On the contrary,
because the robot coexists in a public environment, to determine the overall level of acceptance of innovations, such as
DustCart, it should also be necessary to include ethical, social,
and legal issues.

Conclusions
During the test period, the robot operated for 47 days
(from 15 June 2010 to 7 August 2010), for a total of 454 h
(with reduced service on Tuesdays), carrying out 382 services and traveling a total of 114.6 km. A total of 560.3 kg of
rubbish was collected (paper: 226.5 kg, 40.42%; plastic: 89.7
kg, 16.01%; and undifferentiated garbage: 244.1 kg, 43.57%).

Acknowledgments
The DustCart robot was originally developed in the
framework of the European Union-funded project
DustBot. The authors are grateful to all the people
involved in the organization and realization of the DustCart test, especially Belvedere S.p.A. and its president
Renzo Macelloni, the mayor of Peccioli Silvano Crecchi,

Figure 9. A curious and interested group of people.

IEEE ROBOTICS & AUTOMATION MAGAZINE

MARCH 2011

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - March 2011