Humanoid robots used for surveillance

ABSTRACT the universities, government efforts, industry labs,

Today is a world of new scientific invention. As and countries that contribute to HR, and a cross
this paper includes the use of humanoid robots in section of the disciplines that contribute to the field,
the field of security. There is a wide range of use of such as human, factors, robotics, cognitive
robot like in restaurant, in spying; in medical .our psychology, and design. in the field of defense it is
main objective is to bring this humanoid robot in widely used for searching and surveillance purpose
the field of surveillance. My paper will give because it decreases the human casualties and other
immense use to the army personnel those who are hindrance for the army personnel.
survellancing in the remote areas of the India. Now INTRODUCTION
question arises why only humanoid robots used for Before Da Vinci there was Al-Jazari – the
the surveillance purpose? Now for this answer take Engineering genius of the Islamic world in the
the example of Mumbai terror attack the terrorist middle Ages. He designed and built a number of
have done the damage for 72 hrs.now the N.S.G automatas including the first programmable
doesn’t know how many terrorist are there in hotel humanoid robot. He is also invented the Crank-
Taj .for this reason the robots are designed in a such shank. A 13th Century Programmable Robot team
a way to counter such terror activities . In my paper from the USA history channel was on campus last
I have presented some modification in the robot i.e. month in the Faculty of Engineering to talk about
to use the metal detector, infrared cameras for some very old robots. They were there to film a
sensing in the very dim light or no light replica of the mechanism for al-Jazari’s drinking
.temperature detector for detecting the human inside boat; a boat full of musical automata first
the building. In my paper I have used 10 mega pixel constructed in 1206. Professor Noel Sharkey from
camera 360 rotation to get the clear cut view of the Computer Science built the core of the device
inside building .so this will help the army personnel –”bogged it together from a pile of rubbish”, he
to counter strike with the terrorist. There will be says – to demonstrate how it could have been
fewer casualties to army as well as human being. programmed. The previous claim for the world’s
My motive for bringing this paper to you that there oldest programmable automata is for a machine
is a great scope of enhancement in the field of built by Leonardo da Vinci in 1478.
robotics for the security inside the country. The Al-Jazari’s machine was originally a boat with four
robot will also carry insurgency operation in automatic musicians that floated on a lake to
country for countering terrorism. Human-Robot has entertain guests at royal drinking parties. It had two
recently received considerable attention in the drummers, a harpist and a flautist. Professor
academic community, in labs, in technology Sharkey’s machine has just the one drummer with a
companies, and through the media. Because of this drum, cymbals, bells and no body. The flautist is
attention, it is desirable to present a survey of HR to replaced with an Irish penny whistle. He says he
serve as a tutorial to people outside the field and to wouldn’t risk taking this to any drinking parties
promote discussion of a unified vision of HR within round here.
the field. The goal of this review is to present a The heart of the mechanism is a rotating cylindrical
unified treatment of HR-related problems, to beam with pegs (cams) protruding from it. These
identify key themes, and discuss challenge just bump into little levers that operate the
problems that are likely to shape the field in the percussion. The point of the model is to
near future. Although the review follows a survey demonstrate that the drummer can be made to play
structure, the goal of presenting a coherent "story" different rhythms and different drum patterns if the
of HR means that there are necessarily some well- pegs are moved around. In other words it is a
written, intriguing, and influential papers that are programmable drum machine.
not referenced. Instead of trying to survey every “Whether or not al-Jazari dynamically programmed
paper, we describe the HR story from multiple his machines is an intriguing question”, he says, “it
perspectives with an eye toward identifying themes is quite likely that he used this method, at the very
that cross applications. The survey attempts to least, for fine tuning the rhythm of the musicians”.
include papers that represent a fair cross section of Professor Sharkey is currently looking at a much
older mobile automaton device by Heron of noticeable on power density and system complexity
Alexandria, 1st Century AD, which he now suspects scales. In the first place, most current humanoids
may also have been programmable. [8] aren’t strong enough even to jump and this happens
because the power/weight ratio is not as good as in
the human body. The dynamically balancing Dexter
can jump, but poorly so far. On the other hand,
there are very good algorithms for the several areas
of humanoid construction, but it is very difficult to
merge all of them into one efficient system (the
system complexity is very high). Nowadays, these
are the main difficulties that humanoid robots
development has to deal with. Humanoid robots are
created to imitate some of the same physical and
mental tasks that humans undergo daily. Scientists
HUMANOID ROBOT and specialists from many different fields including
Today in modern era of development of robot has engineering, cognitive science, and linguistics
come in great extent. ROBOT means servant. The combine their efforts to create a robot as human-like
word has been derived from word ROBOTA. as possible. Their creators' goal for the robot is that
A humanoid robot is a robot with its one day it will be able to both understand human
overall appearance, based on that of the human intelligence, reason and act like humans. If
body, allowing interaction with made-for-human humanoids are able to do so, they could eventually
tools or environments. In general humanoid robots work in cohesion with humans to create a more
have a torso with a head, two arms and two legs, productive and higher quality future. Another
although some forms of humanoid robots may important benefit of developing androids is to
model only part of the body, for example, from the understand the human body's biological and mental
waist up. Some humanoid robots may also have a processes, from the seemingly simple act of walking
'face', with 'eyes' and 'mouth'. Androids are to the concepts of consciousness and spirituality.
humanoid robots built to aesthetically resemble a Right now they are used for welding. In the future
human. A humanoid robot is an autonomous robot they can greatly assist humans by welding and
because it can adapt to changes in its environment mining for coal. There are currently two ways to
or itself and continue to reach its goal. This is the model a humanoid robot. The first one models the
main difference between humanoid and other kinds robot like a set of rigid links, which are connected
of robots. In this context, some of the capacities of a with joints. This kind of structure is similar to the
humanoid robot may include, among others: one that can be found in industrial robots. Although
• self-maintenance (like recharging itself)
this approach is used for most of the humanoid
• autonomous learning (learn or gain new
robots, a new one is emerging in some research
capabilities without outside assistance, works that use the knowledge acquired on
adjust strategies based on the surroundings biomechanics. In this one, the humanoid robot's
and adapt to new situations) bottom line is a resemblance of the human skeleton.
• avoiding harmful situations to people,
Enon was created to be a personal assistant. It is
property, and itself self-guiding and has limited speech recognition and
• safe interacting with human beings and the
synthesis. It can also carry things.
environment Humanoid robots are used as a research tool in
Like other mechanical robots, humanoid refer to the several scientific areas.
following basic components too: Sensing, Actuating Researchers need to understand the human body
and Planning and Control. Since they try to simulate structure and behavior (biomechanics) to build and
the human structure and behavior and they are study humanoid robots. On the other side, the
autonomous systems, most of the times humanoid attempt to simulate the human body leads to a better
robots are more complex than other kinds of robots. understanding of it.
This complexity affects all robotic scales Human cognition is a field of study which is
(mechanical, spatial, time, power density, system focused on how humans learn from sensory
and computational complexity), but it is more information in order to acquire perceptual and
motor skills. This knowledge is used to develop Proprioceptive Sensors
computational models of human behavior and it has Proprioceptive sensors sense the position, the
been improving over time. orientation and the speed of the humanoid's body
It has been suggested that very advanced robotics and joints. In human beings inner ears are used to
will facilitate the enhancement of ordinary humans. maintain balance and orientation. Humanoid robots
Although the initial aim of humanoid research was use accelerometers to measure the acceleration,
to build better orthosis and prosthesis for human from which velocity can be calculated by
beings, knowledge has been transferred between integration; tilt sensors to measure inclination; force
both disciplines. A few examples are: powered leg sensors placed in robot's hands and feet to measure
prosthesis for neuromuscular impaired, ankle-foot contact force with environment; position sensors,
orthosis, biological realistic leg prosthesis and that indicate the actual position of the robot (from
forearm prosthesis. which the velocity can be calculated by derivation)
Besides the research, humanoid robots are being or even speed sensors.
developed to perform human tasks like personal Exteroceptive Sensors
assistance, where they should be able to assist the Exteroceptive sensors give the robot information
sick and elderly, and dirty or dangerous jobs. about the surrounding environment allowing the
Regular jobs like being a receptionist or a worker of robot to interact with the world. The exteroceptive
an automotive manufacturing line are also suitable sensors are classified according to their
for humanoids. In essence, since they can use tools functionality.
and operate equipment and vehicles designed for Proximity sensors are used to measure the relative
the human form, humanoids could theoretically distance (range) between the sensor and objects in
perform any task a human being can, so long as the environment. They perform the same task that
they have the proper software. However, the vision and tactile senses do in human beings. There
complexity of doing so is deceptively great. are other kinds of proximity measurements, like
They are becoming increasingly popular for laser ranging, the usage of stereo cameras, or the
providing entertainment too. For example, Ursula, a projection of a colored line, grid or pattern of dots
female robot, sings, dances, and speaks to her to observe how the pattern is distorted by the
audiences at Universal Studios. Several Disney environment. To sense proximity, humanoid robots
attractions employ the use of animatrons, robots that can use sonars and infrared sensors, or tactile
look, move, and speak much like human beings, in sensors like bump sensors, whiskers (or feelers),
some of their theme park shows. These animatrons capacitive and piezoresistive sensors.
look so realistic that it can be hard to decipher from
a distance whether or not they are actually human.
Although they have a realistic look, they have no
cognition or physical autonomy.[9]
Humanoid robots, especially with artificial
intelligence algorithms, could be useful for future
dangerous and/or distant space exploration
missions, without having the need to turn back
Arrays of tactels can be used to provide data on
around again and return to Earth once the mission is
what has been touched. The Shadow Hand uses an
completed
array of 34 tactels arranged beneath its polyurethane
THEORY SENSORS
skin on each finger tip.[2] Tactile sensors also
A sensor is a device that measures some attribute of
provide information about forces and torques
the world. Being one of the three primitives of
transferred between the robot and other objects.
robotics (besides planning and control), sensing
Vision refers to processing data from any modality
plays an important role in robotic paradigms.
which uses the electromagnetic spectrum to produce
Sensors can be classified according to the physical
an image. In humanoid robots it is used to recognize
process with which they work or according to the
objects and determine their properties. Vision
type of measurement information that they give as
sensors work most similarly to the eyes of human
output. In this case, the second approach was used.
beings. Most humanoid robots use CCD cameras as
vision sensors. Sound sensors allow humanoid
robots to hear speech and environmental sounds, problem relies on a major concept, the Zero
and perform as the ears of the human being. Moment Point (ZMP).
Microphones are usually used for this task.[7] Another characteristic about humanoid robots is that
ACTUATORS they move, gather information (using sensors) on
Actuators are the motors responsible for motion in the "real world" and interact with it, they don’t stay
the robot. Humanoid robots are constructed in such still like factory manipulators and other robots that
a way that they mimic the human body, so they use work in highly structured environments. Planning
actuators that perform like muscles and joints, and Control have to focus about self-collision
though with a different structure. To achieve the detection, path planning and obstacle avoidance to
same effect as human motion, humanoid robots use allow humanoids to move in complex
mainly rotary actuators. They can be electric, environments.
pneumatic, hydraulic, piezoelectric or ultrasonic. There are features in the human body that can’t be
Hydraulic and electric actuators have a very rigid found in humanoids yet. They include structures
behavior and can only be made to act in a compliant with variable flexibility, which provide safety (to
manner through the use of relatively complex the robot itself and to the people), and redundancy
feedback control strategies. While electric coreless of movements, i.e., more degrees of freedom and
motor actuators are better suited for high speed and therefore wide task availability. Although these
low load applications, hydraulic ones operate well characteristics are desirable to humanoid robots,
at low speed and high load applications. they will bring more complexity and new problems
Piezoelectric actuators generate a small movement to planning and control.
with a high force capability when voltage is applied. LEARNING AND UNDERSTANDING
They can be used for ultra-precise positioning and BIMANUAL MOVEMENTS
for generating and handling high forces or pressures Scientists at Dublin City University have researched
in static or dynamic situations. a subset of human movements called bimanual
Ultrasonic actuators are designed to produce movements. At different stages of this research they
movements in a micrometer order at ultrasonic have approached the problems from the novel
frequencies (over 20 kHz). They are useful for points of view. They believe that many machine
controlling vibration, positioning applications and learning problems can accommodate neuroscience
quick switching.[6] and perceptual aspects of human movements for
Pneumatic actuators operate on the basis of gas learning and recognizing human behaviors.
compressibility. As they are inflated, they expand Learning and recognizing human movements have
along the axis, and as they deflate, they contract. If been given great attention of researchers around the
one end is fixed, the other will move in a linear world in the recent years. A broad range of
trajectory. These actuators are intended for low applications from medicine to surveillance and
speed and low/medium load applications. Between security can benefit from this technology. Learning
pneumatic actuators there are: cylinders, bellows, hand movements and recognizing gestures are
pneumatic engines, pneumatic stepper motors and significant components of such technologies.
pneumatic artificial muscles. Bimanual movements in general form a large subset
PLANNING AND CONTROL of hand movements in which both hands move
In planning and control the essential difference simultaneously in order to do a task or imply a
between humanoids and other kinds of robots (like meaning. Clapping, opening a bottle, typing on a
industrial ones) is that the movement of the robot keyboard and drumming are some usual bimanual
has to be human-like, using legged locomotion, movements. Sign Languages also use bimanual
especially biped gait. The ideal planning for movements to accommodate sets of gestures for
humanoid movements during normal walking communication.
should result in minimum energy consumption, like Due to the involvement of both hands,
it happens in the human body. For this reason, understanding bimanual movements requires not
studies on dynamics and control of these kinds of only computer vision and pattern recognition
structures become more and more important. techniques but also neuroscientific studies as a
To maintain dynamic balance during the walk, a background to perceive the movements.
robot needs information about contact force and its A cognitive system for bimanual movements
current and desired motion. The solution to this learning and understanding entails three
fundamental components (see Figure 1): low-level where the shapes change. This is especially
image processing to deal with sensory data, important from the processing speed point of view.
intelligent hand tracking to recognize the left hand Since processing and understanding the hands
from the right hand, and machine learning for shapes is usually a time consuming process, as a
understanding the movements. component of an integrated real-time recognition
system, the tracking algorithm must be fast enough
to leave enough room for the other components.
The view-direction and hand-shape independence
naturally lends itself to extending the concept of
A bimanual movement recognition tracking towards mobile vision environments (eg
system. active vision in robotics). We have developed a
model to make the algorithm independent from the
actual position and velocities. Consequently, it can
be used in applications where the visual system (the
camera) moves or turns. For example, assuming that
the camera is installed on a humanoid robot, the
algorithm tracks the hands of a subject while the
robot walks.[3]
The third component of the system is the
recognizer. As a hierarchical cognitive system, it
analyses the hand shapes at the bottom level, learns
The recognition system. the individual partial movement of each hand at the
Using a monochrome surveillance CCD camera the intermediate level, and combines them at the top
hands are extracted based on the hand grey-levels level to recognize the whole movement (see Figure
within the high contrast images. The second 2). Statistical and spatio-temporal pattern
component is hand tracking, which is a significant recognition methods such as Principal Component
problem due to the presence of hand-hand Analysis and Hidden Markov Models form the
occlusion. When one hand covers the other partially bottom and intermediate levels of the system. A
or completely, they must be re-acquired correctly at Bayesian inference network at the top level
the end of the occlusion period. [5] perceives the movements as a combination of a set
Studies in neuroscience show that the two hands are of recognized partial hands movements.
temporally and spatially coordinated in bimanual The recognizer has been developed so that it learns
movements. In addition, the components of one single movements and recognizes both single and
hand are temporally coordinated too. These concatenated periodic bimanual movements. The
coordination’s form the basis of our algorithm to concatenated periodic bimanual movements are
track the hands in bimanual movements. used particularly in Virtual Reality simulators for
We have taken a general view of the tracking interacting with virtual environments. A virtual
problem to cover many challenging problems in this spacecraft controlled by bimanual gestures is an
area. For example, from a pure pattern recognition example. In all parts of this research we have
point of view a movement can be understood looked at the problems from the general point of
differently when it is seen from different camera view and developed general solutions. The tracking
view directions. By defining a general set of algorithm can be employed in a wide range of
movement models independent of view angle we applications including recognition, Virtual Reality,
have developed the tracking algorithm so that it and surveillance/security systems. The recognizer
covers almost every camera view direction. It is can be used in recognizing both single and
trained in just one direction and can be used in other concatenated periodic bimanual movements.
directions. This makes the algorithm independent of Our plan for the future is to make the recognition
the position of the visual system. component independent from the camera view
Using the temporal coordination’s both between direction. This will result in a system that can
limbs (the two hands) and within a limb (a hand and recognize the movements from the view directions
the fingers) the algorithm tracks the hands that has not been trained for. Results of the ongoing
independent of the hand shapes even in movements research in this area will open significant doors
towards the general learning and understanding of Symposium on Human Factors in Driver
human movements. Assessment, Training and Vehicle Design.
COMPONENT ASSEMBLED IN ROBOT
WEB CAM: This web cam will be mounted on the 2 [2] Proceedings of the IEEE Conference on
top of the robot and it is free to rotate in 360 Intelligent Transportation Systems.
degrees in any direction. Webcams typically include 3 [3] European Annual Conference on Human
a lens, an image sensor, and some support Decision Making and Control.
electronics.
4 [4] J. A. Adams, "Human-robot interaction
design: Understanding user needs and
requirements," in Human Factors and
Ergonomics Society 49th Annual Meeting, 2005.
Various lenses are available, the most common in 5 [5] J. A. Adams and M. Skubic, "Special issue on
consumer-grade webcams being a plastic lens that human-robot interaction," IEEE Transactions on
can be screwed in and out to set the camera's focus. Systems, Man, and Cybernetics: Part A --
Fixed focus lenses, which have no provision for Systems and Humans, vol. 35, no. 4, 2005.
adjustment, are also available. As a camera system's
6 [6] AICHI, Robot Project: We Live in the Robot
depth of field is greater for small imager formats
Age at EXPO 2005. Available from:
and is greater for lenses with a large f-number
http://www.expo2005.or.jp/en/robot/robot_projec
(small aperture), the systems used in webcams have
t_00.html, 2005.
sufficiently large depth of field that the use of a
fixed focus lens does not impact image sharpness 7 [7] J. S. Albus, "Outline for a theory of
much. Image sensors can be CMOS or CCD, the intelligence," IEEE Transactions on Systems,
former being dominant for low-cost cameras, but Man, and Cybernetics, vol. 21, no. 3, pp. 473-
CCD cameras do not necessarily outperform 509, 1991.
CMOS-based cameras in the low cost price range.
Most consumer webcams are capable of providing 8 [8] S. All and I. Norbaksh, "Insect telepresence:
VGA-resolution video at a frame rate of 30 frames Using robotic tele-embodiment to bring insects
per second. Many newer devices can produce video face-to-face with humans," Automous Robots,
in multi-megapixel resolutions, and a few can run at Special Issue on Personal Robotics, vol. 10, pp.
high frame rates such as the PlayStation Eye, which 149-161, 2001.
can produce 320×240 video at 120 frames per 9 [9] R. O. Ambrose, H. Aldridge, R. S. Askew, R.
second.[4] R. Burridge, W. Bluethmann, M. Diftler, C.
2) METAL DETECTOR: metal detector is mounted Lovchik, D. Magruder, and F. Rehnmark,
on the lower part of robot so that it can detect all the "Robonaut: NASA's space humanoid," IEEE
guns and other unwanted metals used by the Intelligent Systems and Their Applications, vol.
terrorist in the process of hostages etc. 15, no. 4, pp. 57-63, 2000.
CONCLUSION
Now today is the day of safety. Every country wants
to be safe in all modes of terror attack inside or
outside the country. My aim for presenting this
paper is to tell you about how good these humanoid
robots can carry their task without any harm to the
human. It decreases the human effort and carries
their job in very best manner. I would like to take
notice of terror attack that has occurred in
Mumbai .the N.S.G commandos has suffered great
set back in their operation as all the operation is
performed manually by the commandoes.
REFERENCES
1 [1] Proceedings of the International Driving