Alive Human Body Detection system using an

Autonomous Mobile Rescue Robot

Sandeep Bhatia Hardeep Singh Dhillon Nitin Kumar
Assistant Professor, ECE Deptt. Assistant Professor, ECE Deptt. Assistant Professor, ECE Deptt.
Raj Kumar Goel Institute of Tech. Punjab College of Engg. & Tech. Sunderdeep Engineering College
Ghaziabad (U.P) India Lalru Mandi, India Ghaziabad (U.P) India
[email protected] [email protected] [email protected]

Abstract— In this paper, a new approach for detecting alive life and property various essential services (like fire
humans in destructed environments using an autonomous brigades, medical and paramedical personnel, police) are
robot is proposed. Human detection in an unmanned area deployed. Some lose their lives because of not being
can be done only by an automated system. Alive human treated at time. According to the field of Urban Search
body detection system proposed a monitoring system using
and Rescue (USAR), the probability of saving a victim is
ultrasonic sensors and camera to record, transmit and
analyze conditions of human body. The task of identify high within the first 48 hours of the rescue operation,
human being in rescue operations is difficult for the robotic after that, the probability becomes nearly zero. Generally,
agent but it is simple for the human agent. In order to detect Rescue People cannot enter into some parts / places of the
a human body, an autonomous robot must be equipped with war field or in the earth quake affected areas. All of these
a specific set of sensors that provide information about the tasks are performed mostly by human and trained dogs,
presence of a person in the environment around. This work often in very dangerous and risky situations. To avoid
describes an autonomous robot for rescue operations. The such losses, a robotic system can perform well for
proposed system uses an ultrasonic sensor in order to detect providing alert (detection) of human being.
the existence of living humans and a low-cost camera in
order to acquire a video of the scene as needed. Additional,
other sensors include temperature, fire and metal detector The main purpose of the robot is to detect alive human
works as bomb sensor to detect the presence of bomb in beings after the occurrence of natural calamities with the
Warfield and in rescue operations. Having detected a sign of help of Ultrasonic sensor. The robot based system will
a living human, the ultrasonic sensor Triggers the camera to sense the radiation of human being and condition the
show live scene. The video is then displayed on the screen. sensed signal to communicate to the control section of
This approach requires a relatively small number of data to this robot. Based on the responded commands the robot
be acquired and processed during the rescue operation. This will react upon. The rescuer may become a victim who
way, the real-time cost of processing and data transmission needs to be rescued. This is why since some years mobile
is considerably reduced. This system has the potential to
robots have been proposed to help them and to perform
achieve high performance in detecting alive humans in
devastated environments relatively quickly and cost- tasks that neither human, dogs nor existing tools can do.
effectively. The detection depending on a number of factors For this project, we will focus only on robots which will
such as the body position and the light intensity of the scene. work in a disaster environment of manmade. The
Results show that the system provides an efficient way to proposed system uses an ultrasonic sensor in order to
track human motion. The aim of this article is to present our detect the existence of living humans and a low-cost
experience with various sensors designed and developed. camera in order to capture video of the scene as needed.
Having detected a sign of a living human, the ultrasonic
Keywords-Autonomous mobile robot, Human Detection, sensor triggers a camera to capture a video of the scene.
Motion Detection, Mobile Robot, Navigation, Rescue,
Robotics, ultrasonic sensor, Urban Search and Rescue.
The simulated robot is assumed to have the capability to
determine its current location in real-time, to wirelessly
communicate with the rescue team, and to locally store
I. INTRODUCTION the status and location information about the trapped
Disasters can disrupt economic and social balance of the victims in case the wireless communications link is
society. Natural disasters occur frequently nowadays. temporarily disconnected.
Many human beings are victims of such occurrences.
Because of high rise buildings and other manmade While detecting human being, there may be some
structures urban and industrial areas can be consider to be obstacle on the way of robot. In order to identify the
more susceptible to disasters. These disasters can be obstacle and to have an alternate path to perform its
categorized into natural and human induced disasters. defined task an ultrasonic detection is used which
Natural disasters include floods, storms, cyclones, effectively fulfills this operation.
bushfires and earthquakes where as besides natural Traditionally, equipped robots faced a lot of problems:
disasters, the urban environment is prone to human i) A very high communication cost was incurred due to
induced disasters such as transportation accidents, the large number of transmitted images to the operator.
industrial accidents and major fires. During such ii) More than 25% of the communications between
calamities, especially disasters, in order to prevent loss of Wireless robot and the control unit was extremely noisy
and therefore was useless. This eventually led to a loss of the scene were acquired and subtracted from each other to
communication between the robots and the operator. As a discover if a motion has occurred. The used technique
result, the robots stopped working totally. was fairly efficient in detecting the victims. However, the
iii) A very high processing cost is also incurred by the robot was not fully autonomous and was dependent on
Capturing, storing, and transmitting a large number of the operator.
images.
iv) Third, there was a continuous need for illumination Gas analysis based bio-sensors [6] CO2 sensors allow
due to the dark nature of these environments. This to detect the carbon dioxide emission, and even the
requires a relatively large power supply, which is not breathing cycle of a victim. It is thus possible to
feasible in such situations. determine if he is still alive but the disadvantage is that
the response time of a CO2 sensor is very slow and that
the sensor has to be very close to the victim to have
II. RELATED WORK useful data because it is very directional and depends
In this section, a brief discussion of some of the related much on the air conditions like humidity, temperature,
work is presented; focusing on the used approach and its wind, and dust. This makes it difficult to use it in a
advantages and disadvantages. Remote Operated and disaster area so this approach was not pursued in this
Controlled Hexapod (ROACH): [12] ROACH is a six project. Like the CO2 sensor, SpO2 is used to determine
legged design that provides significant advantages in if somebody is still alive. It needs direct contact with the
mobility over wheeled and tracked designs. It is equipped person. SpO2 sensing of blood oxygen content requires
with predefined walking gaits, cameras which transmit only the pressing of a small, cool, red light on any bare
live audio and videos of the disaster site, as well as inch of skin. It is not a common sensor in robotics. This is
information about locations of objects with respect to the not a commercially available sensor.
robot’s position to the interface on the laptop. Kohga:
University of Tokyo - The most complicated task for Greer, Kerrow, & Abrantes 2002, [9] represented a
most of the USAR robots has been working on a rough thorough understanding of the urban disaster environment
terrain. Specialized robots have been designed for these and an appreciation for traditional search and rescue
types of environments such as KOHGA the snake like techniques are crucial to determining the success of a
robot. The robot is constructed by connecting multiple hovering robot solution. In this paper search and rescue
crawler vehicles serially, resulting in a long and thin environment is described, the applications of robots in
structure so that it can enter narrow space. urban search and rescue, an outline of robotic
competitions in simulating a real rescue environment is
Quality work has been done in the field of robotics. described.
These robots came into existence in the early 21st century
but since then enormous improvements have been made Bahadori [16] presents an analysis of techniques that
in the concept, design owing to which their capabilities have been studied in the recent years for human body
have improved significantly. Various rescue robots have detection (HBD) via visual information. The focus of this
been developed and some of these are – CRASAR work is on developing image processing routines for
(Centre for Robot-Assisted Search and Rescue): autonomous robots operating for detecting victims in
University of South Florida. This robot[10] was used for rescue environments. The paper both discusses problems
first time in real conditions on 11th September 2001 in arising in human body detection from visual information
the World Trade Centre disaster. Different sensors like and describes the methods that are more adequate to be
millimetre wave radar for measuring distance, a colour applied in a rescue scenario. Finally, some preliminary
CCD camera for vision and a forward-looking infrared experiments for such methods in recognizing rescue
camera for the human heat detection are used in it. victims are reported.
Pissokas [1] describe the social impact of urban
Burion [6] presented a project that aims to provide a devastations has given rise to the field of Urban Search
sensor suite for human detection for the USAR robots. and Rescue Robotics. The aim of this article is to present
This study evaluated several types of sensors for our experience and experimental results with various
detecting humans such as pyroelectric sensor, USB sensors designed and developed.
camera, microphone, and IR camera. The pyroelectric
sensor was used to detect the human body radiation, but III. BLOCK DIAGRAM
its limitation was its binary output. The USB camera was As shown in fig 1. This system consists of transmitter and
used for motion detection, but its limitation was its receiver unit. Furthermore transmitter unit consist of
sensitivity to changes in light intensity. The microphone rotational unit which has a specific set of sensors mount
was used for long duration and high amplitude sound on it and a microcontroller. Receiver unit consist of
detection, but it was severely affected by noise. Lastly, camera which gets activated and starts capturing images
the IR camera was used to detect humans by their heat as soon as ultrasonic sensor detects motion of human
image, but it was affected by other nearby hot objects. being and buzzer gets a beep and there is also an LCD
The main idea was to detect a change in the image scene which is used to display the temperature, motion and
by checking the values of the pixels. Several images for metal detection.
 be changed by making necessary changes in the program
developed.

Start

Navigate

Ultrasonic sensor
checks for Motion

No

Ultrasonic signal?

Others sensors
Yes gets activated

Buzzer gives a beep

Fig. 1 Human Body Detection System Based on Sensors

IV. FLOW CHART

As shown in fig 2, In the flow chart of developed system Stop
first the Robocar navigate in a open field and ultrasonic
sensor checks for motion if yes then buzzer gets a beep
and camera gets activated and acquire an image, at this Camera gets activated
time others sensors get activated, this will show the
presence of human being and transmitter unit will send
the current location and information to the receiver end, if Acquire an image
No then Ultrasonic sensor again checks for motion. If
receiver unit detects motion then robocar takes 900 turn
either left or right. Process an image

V. HARDWARE RESULTS
Manually Capturing is started
In order to simulate the robot, a laptop used T.V Tune
Card as well as LCD to display data interfaced wirelessly
to the Robot mechanism. A Robot is equipped with
following sensors and camera: No
Human
i) Bomb (Metal) sensor to detect the presence of Detected? Yes
suspected material in Rescue operations.
ii) An ultrasonic sensor to detect human motion and
obstacles come in the way of Robot. Yes
iii) Temperature and fire sensor to measure temperatures
and display them on LCD in the form of three states Send location and information to rescue team
High, Low and Medium.
iv)Camera to record and display data when sensor
triggers it.
According to the hardware developed, the buzzer beeps Done
as soon as ultrasonic sensor detects motion. The robocar
which otherwise follows a particular pattern stops for 8-
Stop
10 seconds as soon as motion is detected. After some
delay when there is no motion the robocar again follows
the same pattern. The pattern which robocar follows can Fig 2. Flow chart of the Developed System
A. Circuit Diagram of Transmitter unit using
ORCAD

Hardware of Transmitter and Receiver Unit

Fig. 5 & 6. Complete Hardware for the system

Fig. 3 Transmitter Circuit Diagram

Fig. 7. Motion, Temperature and Metal Detection display on LCD

B. Circuit Diagram of Receiver unit using ORCAD VI. TRANSMITTER END PROTOCOL
Hurdle Detected……………..P2.1 (Active high)
Metal Detected……………….P2.0 (Active low)

Temperature sensor
Limit 1 ……………………….P2.2
Limit 2 ………………………..P2.3 Both are active high

RF sending
10 P3.0
11 P3.1
12 P3.2
13 P3.3
H-Bridge
14 P3.4
15 P3.5… p (pins) All are connected to MCUunit
16 P3.6
17 P3.7
Relay P1.0
Fig. 4 Receiver Circuit Diagram LED P 1.1
Protocol
HARDWARE DEVELOPED D3 D 2 D 1 D 0
0 0 Temp. Low
0 1 Medium………………..L1
1 0 Hot temperature……….L2

0 No hurdle
1 Motion Detected
0 No metal
1 Metal detected

VII. SOFTWARE RESULTS

When the ultrasonic sensor detects a signal, the control
program orders the camera to display the surrounding
area. If a human is detected, the system sends its current wide range of sensors have been developed, but there are
location to the rescue team in addition to the video many problems faced by them such as cost, size,
display if needs be. Figure 2. Shows the flowchart of the environment difficulties etc.
system operation.
REFERENCES
The Robot also has a wireless RF Transmitter and sends [1] P i s s o k a s , J o h n a n d M a l c o l m , C h r i s “ Experiments with
the message to the Remote Location whenever it finds Sensors for Urban Search and Rescue Robots” Dept. of Computer
any alive human. As soon as motion is detected, data is Science, University of Essex a n d Division of Informatics,
University of Edinburgh, Scotland UK, 2001..
taken serially through serial port. Since the sensor has a
[2] Rufaida Shamroukh, Fahed Awad “Detection of surviving humans
digital output it sends logic 1 for motion detected and in destructed environments using a simulated autonomous robot”
logic 0 for no motion. As soon as logic 1 is sent, i.e. IEEE Transaction, March 24-26, 2009.
according to the program developed ‘a’ is transmitted [3] Carlos Marques, Pedro Lima, Rodrigo Ventura “Semi-autonomous
robot for rescue operations” Proceedings of the 2006 IEEE/RSJ
serially to the PC, the webcam gets activated and International Conference on Intelligent Robots and Systems,
manually capturing is activated. October 9 - 15, 2006, Beijing, China.
[4] R. Fuksis, M. Greitans, E. Hermanis “Motion Analysis and
Assembly language programming has used in order to Remote Control System using Pyroelectric Infrared Sensors”
IEEE, 2008. No. 6(86), PP. 69-72.
simulate the robot. We can also use MATLAB in order to [5] Jean Schultz, Jill L. Drury, Holly A. Yanco “Evaluation of
operate robocar. Human-Robot Interaction Awareness in Search and Rescue”
IEEE, 2004, PP. 2327-2332.
VIII. APPLICATIONS [6] Steve Burion “Human Detection for Robotic Urban Search and
Rescue” Info science database of the publications and research
• In military applications to detect the presence of reports. Technical Report, February, 2004. Record number 29999.
human being. [7] Kun-Mu Chen, Yong Huang, Jianping Zhang and Adam Norman
“Microwave Life-Detection Systems for Searching Human
• In Rescue operations where human reach is not Subjects Under Earthquake Rubble or Behind Barrier” IEEE
possible. Transactions, Biomedical Engineering, VOL. 27, NO. 1, January
2000, PP. 105-114.
• In Medical applications to detect motion. [8] Binoy Shah and Howie Choset. "Survey on Urban Search and
Rescue Robotics" Journal of the Robotics Society of Japan,
• In Warfield affected areas, to detect the presence Vol.22, No.5, Page.582-586, 2004.
of bomb. [9] Greer, D., Kerrow, P. M., & Abrantes, J. “Robots in urban Search
and Rescue Operations” Australasian Conference on Robotics and
Automation, Auckland, 27-29 November, 2002, PP. 25-30.
IX. CONCLUSION [10] Casper, J. “Human-Robot Interactions during the Robot-Assisted
The goal of this research was to provide a low cost rescue Urban Search and Rescue Response at the World Trade Center”,
IEEE, 2003, PP.367-385.
robot for human detection in a disaster environment. [11] Afak Sekmen, Ali S., Kawamura, Kazuhiko, and Wilkes, Mitch
Though, the existing Urban Search and Rescue Robots “An Application of Passive Human–Robot Interaction: Human
are equipped with various sensors, but the problem with Tracking Based on Attention Distraction” IEEE, Vol. 32, No. 2,
them is the cost. The sensors used in the development of March 2002, PP. 248-259.
[12] Amerada, T., Yams, T., Igarashit, H., & Matsunos, F.
this project are easily available and cost effective. “Development of the Snake-like Rescue Robot KOHGA” IEEE,
2004, PP. 5081-5086.
In this paper, a new method for detecting surviving [13] Ying-Wen Bai, Li-Sih Shen and Zong-Han Li “Design and
humans in destructed environments using simulated Implementation of an Embedded Home Surveillance System by
Use of Multiple Ultrasonic Sensors” IEEE Transactions on
autonomous robot is proposed. The robot uses two levels Consumer Electronics, Vol. 56, No. 1, February 2010, PP. 119-
of sensing in order to achieve higher cost-effectiveness in 124.
the detecting process in terms of the actual cost of [14] Kai-Rui Zhao, Xin-Min Wang, Van Li, and Xiang Yu, "A Life-
equipment, the processing cost, the communication cost, detection System for Special Rescuing Robots" IEEE, 2006.
[15] Miyama, S., Imai, M. & Anzai, Y. (2003). “Rescue Robot under
the storage cost, and the power cost. The first level is an Disaster Situation: Position Acquisition with Omni-directional
ultrasonic sensor that is used as the primary sensor in Sensor”, in Proceedings of the 2003 IEEERSJ Int. Conference on
order to detect the existence of living humans in a scene. Intelligent Robots and Systems, Las Vegas, Nevada, October.
The second level is a human body shape sensor. This [16] Bahadori, Shahram., Iocchi, Luca. “Human Body Detection in the
RoboCup Rescue Scenario” Department of informatics, Rome
level uses low-cost web camera in order to confirm the Italy.
existence of a human shape. The robot is assumed to be [17] Roger, B., Hong, T., Madhavan, R., & Weiss, B, “3D Range
equipped with a simple Temperature and bomb sensor in Imaging for Urban Search and Rescue Robotics Research”, in
order to detect fire in Rescue scenario and suspected Proceedings of the 2005 IEEE, International Workshop on Safety,
Security and Rescue Robotics, Japan, PP. 164-169, June 2005.
metal respectively and a wireless communication link in [18] Hong liu, Ning dong, Hongbin zha “Omni-directional Vision
order to communicate with the rescue team whenever a based Human Motion Detection for Autonomous Mobile Robots”
need arises. IEEE, 2006.
[19] Song, Byunghun., Choi , Haksoo and Lee, Su, Hyung,
“Surveillance Tracking System Using Passive Infrared Motion
X. FUTURE SCOPE Sensors in Wireless Sensor Network” IEEE, PP. 463-816.
Since the system developed is a low cost system therefore
it has a wide future scope. Though many systems with a