Beruflich Dokumente
Kultur Dokumente
SEMINAR REPORT
submitted by
AJAYGOSH P.S.
Reg. No. REANEEC007
Eighth Semester
BACHELOR OF TECHNOLOGY
in
Electronics and Communication Engineering
UNIVERSITY OF CALICUT
March 2016
Department of Electronics and Communication Engineering
Royal College of Engineering and Technology
Akkikavvu , Thrissur -680 604
( www.royalcet.ac.in)
Bonafide Certificate
This is to certify that the seminar report entitled “CYBORG INTELLIGENCE:RECENT
PROGRESS AND FUTURE DIRECTIONS”, is a bonafide record of work carried
out by AJAYGOSH P.S. (Reg.No.REANEEC007) of Eighth Semester Electronics and
Communication Engineering (2013 admissions) of Royal College of Engineering and
Technology, as a requirement for the award of Degree of Bachelor of Technology in
Electronics and Communication Engineering of UNIVERSITY OF CALICUT dur-
ing the academic year 2016-2017.
Mr.GNANAJEBADAS Dr.V.Vijayarangan
Asst. Prof., Dept. of ECE Dept. of ECE
Acknowledgement
First and foremost I thank God Almighty for His providence and for being the
guiding light throughout the seminar.
I want to start expressing my thanks to my seminar guide Mr.GNANAJEBADAS
, Asst. Prof., Dept. of Electronics and Communication Engineering, because of her
valuable advice and guidance towards this work. I received motivation, encouragement
and hold up from her during the course of work.
Finally, I would like to extend my sincere gratitude to friends who have always
been helpful, in preparing and presenting the report and in the discussion following the
presentation.
AJAYGOSH P.S.
Reg. No. REANEEC007
Eighth Semester
Electronics and Communication Engineering (2013 Admissions)
Royal College of Engineering and Technology
Akkikavvu , Thrissur -680 604
March 2016
ABSTRACT
Contents
ACKNOWLEDGEMENT i
ABSTRACT ii
LIST OF FIGURES iv
1 INTRODUCTION 1
1.1 Report Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 LITERATURE SURVEY 3
5 CONCLUSION 24
5.1 Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.2 Disadvantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3 Future scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
REFERENCES 26
List of Figures
2.1 Three main components of our rat cyborg system. The electrode picture
is taken under a microscope. The rat-mounted pack includes a miniature
camera, a wireless module, and a stimulator. . . . . . . . . . . . . . . . . 7
List of Abbreviations
AI Artificial Intelligence
ECoG Electrocorticogram
CO Cyborg Olympics
Chapter 1
INTRODUCTION
Chapter 2
LITERATURE SURVEY
1. Author: Yueming Wang, Minlong Lu, Zhaohui Wu, Liwen Tian,Kedi Xu and Xi-
aoxiang Zheng,Gang Pan
Title: Visual Cue-Guided Rat Cyborg for Automatic Navigation
Published in: , IEEE Computational Intelligence, vol. 10, no. 2, 2015, pp. 4252.
objects in front of it until the rat completes the motion successfully. Thus, the rat
robot, which we refer to as a rat cyborg, is able to move according to the detected
objects without the need for manual operations. The object detection methods and
the closed loop stimulation model are evaluated in experiments, which demonstrate
that our rat cyborg can accomplish human-specified navigation automatically.
A rat robot is a typical animal robot, which can navigate along a hu-
manspecified route. A major disadvantage is that humans need to identify the ar-
rangements of objects in the environment before giving appropriate instructions to
facilitate navigation in the environment. This limits the possible applications of rat
robots in environments that humans cannot observe. In some applications, only a
few objects are of interest to a rat robot, such as human faces or indication signs.
If the rat robot system can find these objects and a motion action is specified for
each object, this would allow the rat robot to perform human-specified navigation
automatically. In this preliminary study, we attempt to address this problem. We
construct a rat robot where the rat can accept stimuli and perform a few basic ac-
tions, such as turning left, turning right, and walking forward. Our novel system
makes two major contributions, as follows.
1) To allow the rat robot to find human-interesting objects, i.e., the objects that eas-
ily attract humans attentions such as human faces and indication signs, a miniature
camera is mounted on the back of the rat robot and the video captured by the cam-
era is transferred to a computer. Interesting objects in the video, such as human
faces and arrow signs, are then identified by object detection Visual algorithms and
2) To allow the rat robot to navigate automatically while being guided by the iden-
tified objects/cues, we develop a stimulation model that drives the rat robot to per-
form a unique motion action in response to the detection of an object. A problem
with automatic control is that a single stimulus, e.g., a stimulus for turning left,
does not allow the rat to perform a successful turning left motion. Humans usually
give a series of stimuli to the rat for this purpose, according to the state of the rat
and the objects in front of it. Inspired by this manual control process, we develop
a closed-loop stimulation model that mimics the human control procedure, which
issues a stimulus sequence automatically according to the state of the rat and the
objects detected until the rat completes the motion successfully.
The first aerial animal robot was described in 1997, where the locomo-
tory reaction of a Periplaneta americana in response to various electrical stimuli
was analyzed. Using two photosensors as inputs, an electronic backpack was used
to drive the insect to walk along a black line. The success of the Periplaneta amer-
icana robot greatly inspired research into animal robots based on insects. Micro-
probes were implanted in Manduca sexta during its early metamorphosis to allow
direct control of its wing motions, thereby controlling the flight direction in Man-
duca sexta.
pared.
Terrestrial animal robots such as Gekko gecko and rats, have also been
investigated. A rat robot was first developed in, where applications of electrical
stimuli to the somatosensory cortices (SI) and medial forebrain bundle (MFB) were
used as cues and rewards, respectively. The rat could easily be guided through pipes
and across elevated runways, and it could even be instructed to climb or jump from
trees. A new method for rat robot navigation was proposed in, which was based on
virtual punishment.
A new locomotion control scheme was developed for rat navigation. Al-
though animal robots have been studied extensively, few efforts have been made to
develop an animal robot that is capable of performing automatic navigation. In this
study, we incorporate object detection algorithms in a rat robot and the objects iden-
tified are used to guide automatic navigation via a closed-loop stimulation model.
Figure 2.1: Three main components of our rat cyborg system. The electrode picture is
taken under a microscope. The rat-mounted pack includes a miniature camera, a wireless
module, and a stimulator.
The future of cyborg intelligence may lead towards many promising ap-
plications such as neural intervention, rehabilitation, medical treatment, and early
diagnosis of some neurological and psychiatric disorders. It may replace, repair,
assist, and augment human sensory-motor or cognitive functions. For example,
neuroprosthetics can replace a missing body part and still interface with the hu-
man nervous system and brain to increase precision and comfort of movements, or
memory chips for restoring and enhancing memory function. Cyborg intelligence
has the potential to make the bionic man reality. While cyborg intelligence has
many potential exciting applications, research in this area is still in the preliminary
stages.
systems could greatly benefit from biological intelligence, solving problems that
are still beyond the capabilities of the state of the art. For instance, image under-
standing is a relatively easy job for humans, yet it still challenges even the most
sophisticated AI algorithms.
cognitive abilities with the computer systems computational power. The term cy-
borg was coined by Manfred Clynes and Nathan Kline in 1960,3 to describe a
being with both organic and synthetic parts. More broadly, cyborgs refer to sym-
biotic biological-machine systems, consisting of both organic and computing com-
ponents. Cyborg intelligence is a new research paradigm, aiming to combine the
best of both machine and biological intelligence.
Chapter 3
For the sensorimotor process, our previous work abstracted the biological
component of cyborg intelligence into three layers: perception and behavior, decision
making, and memory and intention. We also divided the AI functional units into three
corresponding layers: sensor and actuator, task planning, and knowledge base and goal
layers. We also defined two basic interaction and cooperation operations: homogeneous
interaction (homoraction) and heterogeneous interaction (heteraction). The former repre-
sents information exchange and function recalls occurring in a single biological or com-
puting component, whereas the latter indicates the operations between the function units
of both biological and computing parts. Homoraction is also modeled as the relation-
ship between units within the same part. In the case of a single part in a braincomputer
integration system, it will reduce to a biological body or computing device just with ho-
moraction inside. Consequently, verifying the existence of heteraction is necessary for
cyborg intelligent systems.
We believe that learning and memory units are fundamental for problem
solving in the computational framework of cyborg intelligent systems. Biological learn-
Figure 3.1: Hierarchical conceptual framework for cyborg intelligence.1 Three layers are
abstracted to describe the interconnection between the biological organisms and comput-
ing machines.
ing paradigms (such as classical conditioning, operant learning, and insight learning) and
learning rules (such as sequential and recursive patterns) are essential to generate adap-
tive behaviors, whereas AI algorithms enable computing devices to be smart to perform
intelligent tasks. Cyborg intelligent systems interact with the environment to achieve
better behavior performance, which requires the biological and AI units to adapt to each
other within the system and adjust their actions according to changing environmental sit-
uations, thus possessing enhanced learning, memory, and problem-solving capabilities.
Neural plasticity and machine learning cooperate and merge to represent the cyborg in-
telligent systems learning capacity, usually in a semisupervised or unsupervised manner.
Figure 3.3: Vision-augmented rat cyborg. The ratbot can recognize image indicators
and navigate complicated environments guided by commands from its computing com-
ponents.
surrounding environment. A computer analyzes video stream input and generates stim-
ulation parameters that are then wirelessly sent to the backpack stimulator to control the
rats navigation behavior by manipulating virtual sensation or reward. As Figure shows,
vision enhanced ratbots can precisely find human-interesting objects that is, human faces
and arrow signs, identified by object-detection algorithms.
Figure 3.4: Robotic hand control with neural decoding. We used motor control com-
mands decoded directly from neural spikes of the primary motor cortex to manipulate the
robotic hand to demonstrate four gestures.
alyzed the spatiotemporal patterns of ECoG signals. We extracted the power spectrum
of high gamma frequency components(80 to 120 Hz) to decode rock-paperscissors hand
movements, and we then controlled the prosthetic hand to perform the gestures.
Chapter 4
Chapter 5
CONCLUSION
It seem that in future we may have more and more artificial body parts like
eyes,nose,hands and leg.Also Cyborg system can act as a human body by improving the
quality and the life.
5.1 Advantages
1. There are also brain implants based on neuromorphic modelling .A deaf man has
had his inner ear replaced so he can now engage in telephone conversation. (this in
time will be upgraded so that he can hear music)
2. PARKINSON DISEASE: there are brain implants that help reverse the most dev-
astating symptoms of the disease.
4. Using ratbots Simple commands were sent to them to manipulate their brains and
into following a pre-programmed route/command. This was successful and now
researchers are proposing to use them to search for land mines and buried victims
of earthquakes as they can move more quickly then human rescuers or existing
robots.
5. Prolongs life
5.2 Disadvantages
1. Robots can sense the world in ways that humans cannot-ultraviolet, x-rays, in-
frared and ultra sonic perception. So basically there is more dependence on cyber
technology.
3. Cyborg have no problem thinking of the world around them in wider dimensions
(multiple) where as human beings are more restricted in that sense.
4. Cyborgs have physical limitations. Cyborgs do not heal Body damage normally,
but instead must be repaired. For example, broken limbs and damaged armour
plating must be replaced which can be expensive and time consuming.
References
[1] Zhaohui wu, ”Cyborg Intelligence:Recent Progress and Future Direction”, IEEE
Intelligent Systems 2016.
[3] Y. Wang et al, ”Visual Cue-Guided Rat Cyborg for Automatic Navigation”, IEEE
Computational Intelligence, vol. 10, no. 2, 2015, pp. 4252.
[4] D. Zeng and Z. Wu, ”From Artificial Intelligence to Cyborg Intelligence”, IEEE
Intelligent Systems, vol. 29, no. 5, 2014, pp. 24.