IEEE Robotics & Automation Magazine - December 2010 - 24

Seeing should not imply
believing in the world of
telepresent communication.
set of biometric methods for motion, face, gesture, gait, and
voice identification. However, with the help of advanced
technology, another area appears on the rise that looks into
how computer programs, bots, robots, and avatars could be
detected, identified, and behaviorally profiled by humans.
Examples of situations where such recognition might provide
to be crucial include an unauthorized intelligent bot participating illegally in an online game and pretending to be a real
human, a military autonomous device with a hacked control
system, or a human assistant robot that has been malfunctioning and could become dangerous as a result.
While the theoretical contributions on the subject are of
importance, it is the practical application of the results to realworld problems, as well as emerging areas of computer science,
that are key aspects of this special issue. The four carefully
selected contributions comprising this special issue are connected by a common theme: the application of recent methodological advances to important robotics security-related
problems. Each of the contributions also find a unique niche in
the spectrum of the current research, presenting a new look into
the society of robots and emerging security issues, investigating
authentication problems at the age of increased avatar realism,
presenting a new method for visual robot detection, and introducing an intelligent behavioral model of cognitive robots.
The first article "Toward a Society of Robots: Behaviors,
Misbehaviors, and Security" by Antonio Bicchi, Adriano Fagiolini, and Lucia Pallottino presents a number of new approaches to
improve the overall security of the society of robots. It first considers how very large numbers of robots, differing in their bodies,
sensing, and intelligence, may coexist, communicate, and compete fairly toward achieving their individual goals. Then the
authors discuss some characteristics that the rules defining acceptable social behaviors of robots should possess, as well as possible
threats that may be posed by the misbehaviors of some of its robot
members. The cause of such misbehavior could be due to faults
or malice, and thus, the task to detect and isolate such members
should be carried out through cooperation of peers. The article
presents examples of motion control protocols for arbitrarily large
groups of heterogeneous robots. In particular, some intrusion
detection algorithms, which allow detection of deviance from
society rules as well as algorithms to build a consensus view on
the environment and on the integrity of peers, are introduced.
Many examples support the proposed approach aimed to
improve the overall security of the society of robots.
"The Age of Avatar Realism: When Seeing Shouldn't Be
Believing" by Laurel D. Riek and Robert N.M. Watson
explores a challenging problem of trust that seems natural
24

IEEE Robotics & Automation Magazine

to users of video teleconferencing and robot telepresence.
Authors postulate that recent advances in affect recognition, facial expression synthesis, and speech synthesis are
giving nonexperts access to tools for performing live manipulation of audio and video streams, as well as control signals for
telepresence robots. This, in turn, presents certain security
risks, which regular users might be unaware or unfamiliar
with. Riek and Watson then apply the vocabulary of security
protocol analysis to this new environment, considering how
changes in technology call into question assumptions of trust.
Increasingly, seeing should not imply believing in the world of
telepresent communication, and the authors argue for strong
measures to protect the security of the channels, including
reinforcing traditional security areas (platform and channel
security), and ensuring that proper security protocols are used
wherever they are appropriate.
"Play Ball!: Robot Detection System in the Soccer Domain"
by J. Ruiz-del-Solar, R. Verschae, M. Arenas, and P. Loncomilla addresses the problem of automatic detection and identification of robots, as well as the recognition of their behavioral
patterns. Collaborative robot behaviors in multirobot systems
are becoming highly important in scientific and industrial
application areas, such as ground, space, and underwater exploration, entertainment, surveillance, human assistance, and autonomous rescue operations. A complex variety of interactions
arises among different robots as well as between them and
humans. Thus, the authors argue that it is highly important to
develop automatic, fast, and accurate methods for the detection and identification of robots, as well as the determination
of their behaviors. To address this challenge, the authors present
a new multiclass and multiview robot detection system based
on the trees of cascades classifiers. The successful application of
the proposed methodology in a robotic soccer application
showcases its strengths.
The final article of the special issue, "Cognitive Robots:
A Reference Model Toward Intelligent Authentication" by
Yingxu Wang, introduces an emerging technology of cognitive robots to model the cognitive processes of robot authentication. Fundamental problems remaining in cognitive robot
studies include defining the necessary and sufficient intelligent
behaviors of cognitive robots as well as distinguishing the
intelligent capabilities of cognitive robots from those of their
imperative counterparts. Wang presents a cognitive reference
model of architectures and behaviors of cognitive robots. He
explores cognitive informatics foundations of cognitive robots
from the aspects of neural informatics and abstract intelligence.
The architectural model of cognitive robots is described based
on a layered reference model of the brain. The behavioral
model of cognitive robot is elaborated with the generic behavioral model and the hierarchical relations among the behavioral processes of cognitive robots. Finally, a reference model
of cognitive robots is derived. Wang concludes with applications of the reference model in robot authentication, computational intelligence, and automation systems.
DECEMBER 2010

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - December 2010