NIYANTRA Documents
(Make sure you follow all the 3 steps involved in submission. This is only Step 2 of 3. For details, visit http://bit.ly/16iNfUL)
Contact Information
Name of the College : Meenakshi Sundararajan Engineering College
Name of the Team Members along with their respective current semester : D.Praveen Krishna , 5th Semester
B.Siva Prashanth , 5th Semester
C.Shankar , 5th Semester
E-Mail Address & Phone Number of the Team Leader :krish.achar2112@gmail.com
9500185785
Name of the Faculty Guide :M.Rooba
E-Mail Address & Phone Number of the Faculty Guide : roonand@gmail.com
9940528075
Project Information
Project Title:Automatic Vacuum Cleaner Using Machine Learning
Hardware & Software Used:NI Robotic StarterKit (Version 1 with sbRIO 9631)
BtBee Pro ( Bluetooth Module )
LabView 2012
What challenge/problem are you trying to solve through your application:
Dust cleaning robots are avaliable in markets, which either cleans in an ordered or random way , these robots do not analyse these data or process them. It cleans every point with same point of yiew. It doesent classify the regions of dust density as high or low. This is because the dust occur in the regions where the mobility if the people are least like under the sofa and near windows. This makes the robot to clean the entire house in the same way without thinking about their intensity.
How does your application solves the above mentioned challenge/problem:
Our Project incorporates the learning algorithms( Machine learning) to simplify our daily works.Our Robot is a smart one which cleans the room effectively using machine learning.It classifies the room into two discrete region namely, high and low dust density region , the cleaner cleans the room accordingly by changing the speed tof the cleaner.It increases the cleaning capability of the robot. It is also provided with a feedback loop to rescant the dust ( which is an enhanced version ). Thus we are going into an era of "SMART ROBOTS".
Description of Project:
It creates a test data about the room by measuring the dust density at various points in a random way. These data are then sent to a computer which uses SVM to classify the data and to create a SVM class. The robot is then made to move the room (avoiding obstracles ), it determines its position using the wheel encoding and sends it to the computer . The host VI Process them and determins the dust density using the SVM Class. It is again sent to the robot which controls the speed of the vacuum cleaner. Thus it makes our robot Smarter. There are two main VI for this
1. Host( Server VI)- It is the brain of our robot. In its front pannel we can have the rooms to be selected for cleaning , it also displays the raw data about the dust density from the previous data. It also process the SVM and determine the dust density.
2.Starter Kit VI (in RT) - It basically has a the obstracle avoidance coding in it and also has a coding to determine the position from the wheel encoding. It then sends the data to the computer and receives them.
SVM-
SVM is Support Vector Machine is a type of machine learning which is used to analyze data and recognize the patterns in the given data set. It has a kernel for which we are using gaussian kernel . For more details about machine learning http://en.wikipedia.org/wiki/Support_vector_machine .
Btbee Pro-
The communication is done serially using the Btbee Pro which is a bluetooth module . It is paired with the main computer which facilitates the communication between the computer and the robot.
Insert Images Here:
YouTube Link of Video:
Insert the Video here:\
http://www.youtube.com/watch?v=tBpWhppKmDM&feature=youtu.be
