MILLITARY VEHICLE
S.P.Dhinesh ,V.G.Devika, J.Karthika, M.Pandiyan.
Chennai-602 105
Email: spdhinesh001@yahoo.co.in
ABSTRACT:
Keywords:
The image captured by the webcam is fed into the laptop which is mounted on the
ATV and processed by MATLAB software. It converts the live video into 2D image and
processes it and determines whether it is an obstacles or staircase or corrugations or ramps by
comparing their dimensions. If it finds the obstacles it activates the sensors and follows the
obstacle avoiding algorithm. Figure 2.1 shows the flow diagram for Obstacle Avoidance
Algorithm.
Figure 2.1 Flow diagram for Obstacle Avoidance Algorithm
2.2Navigating in lane
2.3.Staircase climbing
The novelty of this method deals with dynamics of the robot along with
its kinematics and the geometric constrains of the task.This Locomotor was designed with a ground
clearance of 10” to fulfill the functional objective of climbing the stairs. The six wheeled drive
provides superior locomotion and bigger wheels (D 25”) of good tractions with high motor torque
enables the ATV to climb the stairs easily.
ATV uses the GPS system to finds its latitude and longitude position and its
corresponding distance from the waypoints is calculated. Then the map of GPS coordinate system is
converted to cartesian coordinate system.The series of waypoints can also be generated to map the
places. Since the waypoints are in terms of X and Y position, linear paths can be taken. ATV uses
GPS waypoints to make sure that the vehicle is on track.
The basic idea behind the terrain-based navigation is to face the safest, most traversable
region and adjust its speed based on the quality of the terrain to be traversed. The figure 2.2 shows
terrain classification using Image Processing. The motor contoller operates in three different operating
speeds based on the terrain. The various speeds for different terrains is depicted in table 2.1
2.6.E-stop
Wireless E-stop facility is provided to stop the ATV in case of any emergency
situations. This will stop the vehicle at the instant and prevents it from faulty behaviours.
3.Methodology
3.1. Design
The ATV is a six wheeled skid-steer machine using broader tyres as a superior
locomotion to negotiate the three dimensional terrain. The modeling, designing, and
simulation of the dynamics of differential drive is achieved with the aid of ‘Differential Drive
and Global Positioning Blockset’ using MATLAB. The speed of a dc motor is controlled by
PWM method as shown in figure 3.1
3.2System Architecture
The ATV uses Digital image processing which allows a wide range of algorithms
to be applied as input data to Artificial Neural Network (ANN).The ANN decides the course of action
to be taken by the vehicle. The boundary recognition technique is used for lane following and edge
detection for the classification of obstacles. The flow diagram for autonomous navigation is illustrated
in figure 3.3. In the absence of boundary lines the ATV use GPS waypoints to reach the destination.
Fig 3.3: Flow diagram for autonomous navigation
Software Application
MATLAB DIP and ANN
Microsoft Robotics Studio Simulation
GPS Routex GPS navigation
4.Conclusion
Thus the ATV is capable of navigating through different terrains with help of
images obtained by the webcam mounted on the vehicle and traversing through the unknown
environment using the GPS waypoints in a shortest possible time.This ATV can be used in the
army for survailance. These robots will be part of the Low Intensity Conflict / Explosive Ordinance
Disposal (LIC/EOD) program and Homeland Security to carry out tasks into a variety of
undesirable, hazardous, and potentially life-threatening environments in stealth and NBC threats.
5.Future Scope
This ATV can be modified and extended for applications such as Minefield detection &
neutralisation, hazardous material handling and Target identification.
References
1. ‘An Introduction to neural networks‘ by Ben Krose and Patrick van der Smagt.
2. ‘Application of four-layer neural network on information extraction‘; Min Han & Lei Cheng and
Hua Meng.