IEEE Technology and Society Magazine - September 2020 - 42

needs of drivers and other service providers. One
advantage of a CAS solution is that these ethical issues
can be encoded into the fabric of the automation.

Public safety and security involve countermeasures for
protection against threats to the safety and integrity of
the general public, including natural disasters as well
as crimes. CASs may be usefully leveraged for monitoring and for the coordination of responses to dangerous situations.
A motivating example is a distributed surveillance
system, where a collection of cameras are employed for
intrusion detection, target identification and tracking, or
behavioral analysis. The key problems with multi-camera configurations are coordination and control of individual, heterogeneous cameras such that they perform
collaboratively [57]. In other words, the set of cameras
should behave as a CAS in order to react to local and
global contingencies, and to provide distributed global
functions [58].
Regulation in emergency situations is another task
that can be best performed in a collective-adaptive
manner. For instance, coordinating the movement of
people in dangerous situations such as in overcrowded
spaces requires timely, coordinated adaptation, since it
is easy for individual and small-scale deviations to escalate into global chaos, leading to disaster. This problem
is tackled by crowd engineering approaches, which are
often founded on CAS paradigms [32], [59].
Environmental monitoring (discussed in more detail
next in the Ecology section) can also be considered from
the perspective of public safety, as it supports timely
identification and intervention in natural disasters.
However, safety-critical applications pose challenges
for CASs related to verification and validation, real-time
guarantees, and to the satisfaction of "hard" constraints
that are difficult to provably satisfy in highly variable
and possibly unknown scenarios [60]. This remains an
area requiring significant research.

constraints violations (such as hunting of protected animals) or problems detection (such as burning forests).
Moreover most of the adaptation needs of these systems strongly require timely decision making and coordinated and cooperative intervention. Different actors
are involved (including police, firemen, rangers, navy,
national parks, protected reserves, municipalities, environmental managers, and citizens) with diverse physical
distribution and diverse responsibilities in a hierarchy of
managerial roles (for example where a fire brigade unit
intervenes on the spot while headquarters remotely
coordinates several units).
With these parameters, it is easy to see the advantages of using CASs for the overall operation and
dynamic adaptation of systems where heterogeneous
entities can efficiently operate only by collectively sharing strategies and efforts [61]. Furthermore, ecologyoriented systems are also included in the Internet of
Things domain, since smart devices are often exploited in ecology scenarios, which further increases system complexity. For instance, smoke sensors, motion
sensors, GPS, and the like can be used for monitoring
tasks, and even though the IoT might be an enabler of
ecology-oriented systems it might also emphasize the
problems of heterogeneity and interoperability among
the involved agents.
Interesting challenges arise when thinking about the
roles played by different kinds of actors. Smart devices
can act both as passive actors (for example smoke sensors in a monitoring phase) and as active actors during
an adaptation phase (for example drones flying over a
fire to measure its extent and guide rescue efforts). The
same idea applies to humans. Another interesting challenge comes from how and to what extent human
actors and devices cooperate, support, or complement
each other in specific situations. In these contexts,
CASs can represent a proper base from which to investigate dynamic role assignment and communication protocols, taking account of a wide range of diverse actors,
constraints, peculiarities, and uncertainties.

Ecology

Operations Support

The ecology domain offers a wide range of scenarios
that can benefit from the application of collective
approaches and strategies to better handle common
needs. These scenarios span from environmental monitoring (for alerts management in case of wildfires or
floods) to rubbish collection (in the cities as well as the
seas), to the protection of endangered animal species.
Such ecology-oriented systems aim to support the management of medium-to-large areas (e.g., city, sea,
national park) while by guaranteeing efficient operation
during the nominal behavior of systems (such as forest
monitoring), and effective adaptation in case of

One final scenario comes from using CASs to create
resilient, adaptive structures that can aid, sustain, protect, or replace, human activity in dynamic, unknown,
and risky environments. Collaboration between
humans and robots is generally useful in situated problem solving scenarios [62] such as city maintenance or
waste collection.
A particularly relevant case is that of humanitarian
relief operations [63], where people need to be rescued
after environmental catastrophes such as hurricanes,
tsunamis, floods, or during conflict. In these situations,
teams of robots (unmanned ground or aerial vehicles)

Public Safety and Security

IEEE TECHNOLOGY AND SOCIETY MAGAZINE

∕

SEPTEMBER 2020

IEEE Technology and Society Magazine - September 2020

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - September 2020