Poster

GUJARAT TECHNOLOGICAL UNIVERSITY
L. D. COLLEGE OF ENGINEERING
INSTRUMENTATION AND CONTROL ENGINEERING DEPARTMENT
HUMAN ACTIVITY DETECTION ON QUADCOPTER PLATFORM

Shyam V. Chauhan, Fenil M. Parmar, Nishant B. Prajapati and Pathik B. Prajapati.
Guided By. Prof.Chandni Shah
ABSTRACT:

 COMPONENTS:
Block Diagram
PIXHAWK F.C.
Our Project Autonomous Surveillance quad cop- 
ter is flight controller based quad copter which is  TELEMETRY

used for surveillance and monitoring of place or  GPS MODULE
area given by operator. The main features of this  BLDC MOTORS
quad copter are it works automatically and fol-  ESCs
low given path. It is also equipped with High  PROPELLERS
definition camera. After receiving the images or  LIPO BATTERY
video from camera we will use raspberry pi 3 for  FS-I6 R.C.
image processing. It detects the prohibited activi- RASPBERRY PIE
ty happening in the area of surveillance. Many of

CAMERA MODULE
this type of device can be operate from the single 
control room. 
 SOFTWARE:
 MISSION PLANNER
 TENSORFLOW
OPERATION OF PROJECT:
The micro drone here which is a Quad copter is basically mechanism

which is made of 4 symmetric arms with 2 pair of anti directional motors,
which is controlled by some flight controller. The flight controller con-
tains appropriate firmware for control specific components of quad cop-
ter. It gets supply from 3.7 volt 3 cells Li-Po battery. Quad copter can go
in given direction by changing its motor’s direction and speed. In the re-
cent years quad copter is one of the deep projects for research and devel-
opment. ESC stands for Electronic Speed Controller. It converts the PWM
signal from the flight controller or radio receiver, and drives the brushless
dc motor by providing the appropriate level of electrical power. Brushless
DC motor may be described as electronically commuted motor which do
not have brushes. These types of motors are highly efficient in producing PHYSICAL IMAGE
large amount of torque over a vast speed range. Flight controller is the
heart of any quad copter it is not only controls the motors but also it com-
municate with different components and process data and from that it
controls the motors to appropriate direction. It also contains the accel-
erometer and compass.
After successfully planning our mission in quad copter now we added the
key feature of artificial intelligence in our system which is implemented
by Raspberry pi. The Raspberry Pi module is just the CPU and you need
to add the peripherals to it.
We are also adding the pi camera module to capturing live streaming vid-
eo. While we are live streaming the video we simultaneously fed the
stream to python programming algorithm made for artificial neural net-
work algorithm. Artificial neural network in python is developed by the
tensorflow library. Using this artificial neural network programming we
identifying the activity that are restricted in the given area of surveillance.
We are taking over 400 samples of the activity and then fed to the algo-
rithm so it can learn from that.
FUTURE SCOPE & APPLICATION: Acknowledgment:

Our project’s main objective is to take care of the security in the area’s like industries, traffic We express our gratitude to DR. M. T. THAKKER the head of de-
police, Forrest security and many more places. Drone platform is one of the most developing
areas in this field so we equip the latest technology of artificial intelligence and neural net- partment of Instrumentation and Control for his constant encourage-
work for precise result. Due to this product security of the area is constantly observed accu- ment, co-operation and support and also giving the knowledge he
rately and detection of prohibited activity also can be identifying for the better result.
have about our project for making it better.
Most of the drones available today are mainly equipped with manual control through Remote
so it required one special operator to control it. Where if we talk about observing and detect-
we would like to express our deepest sense of gratitude and sincere
ing from the control room also required one special person for identify suspect. thanks to our Project guide PROF. CHANDNI SHAH ma’am for
Our product can be operated automatically and also manually and it is identify the suspect their support and timely co- operation and helping us by giving the
automatically using the latest artificial intelligence technology. Hence the aerial surveillance useful advises for making the project far better than we thought.
becomes more meaningful with less human error.
Reference:
1. Hoffman, G.; Huang, H., Waslander, S.L., Tomlin, C.J., Quad rotor Helicopter Flight Dynamics and Control: Theory and Experiment. In the Conference of the American Institute of
Aeronautics and Astronautics. Hilton Head, South Carolina, 2007.
2. Gowsikhaa D, Manjunath, Abirami S,” Suspicious Human Activity Detection from Surveillance Videos”, International Journal on Internet and Distributed Computing Systems
(IJIDCS), Vol: 2 No: 2, 2012,pp.186-190,2012
3. Emad Samuel Malki Ebeid, Martin Skriver, Jie Jin,” A Survey on Open-Source Flight Control Platforms of Unmanned Aerial Vehicle”, Research Gate Publication, August 2017

Poster

Uploaded by

Copyright:

Available Formats

Poster

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Poster

Uploaded by

Copyright:

Available Formats

GUJARAT TECHNOLOGICAL UNIVERSITY

HUMAN ACTIVITY DETECTION ON QUADCOPTER PLATFORM

ter is flight controller based quad copter which is  TELEMETRY

The micro drone here which is a Quad copter is basically mechanism

FUTURE SCOPE & APPLICATION: Acknowledgment:

You might also like