SMART VACUUM CLEANER

1
SMART VACUUM CLEANER

• PROJECT PRESENTATION
• CSE 591
• PRESENTED BY
• VISHESH MAHAJAN

2
CONTENTS

• INTRODUCTION
• MOTIVATION (WHY)
• PROBLEM DEFINATION (WHAT)
• DESIGN / ARCHITECTURE (WHAT)
• 1. High Level Design
• 2. Software Part
• 3. Basic Algorithm
• 4. Hardware Part
• RESULT (WHERE WHO)
• CONCLUSION, PRESENT WORK, FUTURE WORK (WHEN)

3
INTRODUCTION

• MICROSOFT EASY LIVING
• WE DISUCSSD A SMART ENVIRONMENT, ROOM
• WITH ALL THE FACILITES AND MINIMAL USER
• DISTRACTION.
• BUT EVEN A SMART ROOM NEEDS CLEANING!!!

4
MOTIVATION

• REDUCE USER DISTRACTION
• (TANGENT TYPE DISTRACTION)
• MAKE A COMMON REAL LIFE UTILITY CONTEXT-AWARE
• A

5
PROBLEM DEFINATION

• AIM IS TO CREATE SMART VACUUM CLEANER WHICH HAS
  THE FOLLOWING FEATURES
• 1. AUTOMATIC ON/OFF
• 2. DUST SENSING
• 3. AUTO DUST CLEANING
• 4. AUTO RECHARGING
• A

6
PROB. DEFINATION (CONTD)

• THE VACUUM CLEANER STARTS FROM A CHARGING BASE AT
  SOME PRE-DEFINED TIME (BASED ON USER PREFRENCE)
  AND BASED ON AN ALGORITHM IT SEARCHES, SENSES
  DUST AND SUCKS IT. AFTER CLEANING IT COMES BACK
  TO ITS ORIGINAL POSITION FOR RECHARGING. IT CAN
  PERFORM 4 OPERATIONS
• 1. MOVE FORWARD
• 2. TURN LEFT
• 3. TURN RIGHT
• 4. SUCK

7
ARCHITECTURE

• HLD

(FINAL ROOM STATE)
ACTION
PERCEPT
RECEIVER
(INITIAL ROOM STATE)
SENSOR
COMPUTING ALGORITHM
TRANSMITTER
HARDWARE
INTERFACE
8
SOFTWARE PART

• DUST GENERATION
• RANDOM NUMBER GENERATOR
• BASIC ALGORITHM
• SEARCHING AND PATH GENERATION

9
SOFTWARE PART (CONTD)

• DIVIDES THE ROOM IN A GRID

0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
1
INITIAL ROOM STATE
FINAL ROOM STATE
10
BASIC ALGORITHM

• THE SIMPLEST TECHNIQUE CAN BE TO SEARCH ALL THE
  SQUARES.
• THIS IS NOT AN EFFICIENT STRATEGY.

11
BASIC ALGORITHM

• 3 PRACTICAL OPTIONS
• 1. DEPTH FIRST
• 2. BREADTH FIRST
• 3. ITERATIVE DEEPENING

12
BASIC ALGO (CONTD)

• DEPTH FIRST
• Has modest memory requirements but may get
  stuck going down the wrong path.
• BREADTH FIRST
• It always comes up with the solution but
  takes too much memory and space.
• ITERATIVE DEEPENING
• It combines both DFS and BFS in the sense
  that it is complete as BFS but requires less
  memory as DFS.

13
HARDWARE PART

• THIS DEALS WITH THE INTERFACING OF VACUUM CLEANER
  WITH THE COMPUTER WHICH WILL GIVE COMMANDS TO
  PERFORM THE 4 ACTIONS BASED UPON THE PATH DECIDED
  BY THE SEARCH ALGORITHM.
• THERE CAN BE 2 APPROACHES FOR THIS.

14
HARDWARE PART (CONTD)

• APPROACH 1
• SWITCHING IN THIS COMPUTER WILL GENERATE 1
  SIGNAL AND BASED UPON THE ALGORITHM THE SIGNAL
  WILL BE SIWTCHED BETWEEN 4 OPERATIONS.
• WE WILL HAVE 4 CLOSED CIRCUITS AND THE
  CIRCUIT SELECTED WILL PASS THE SIGNAL TO RECEIVER
  MOUNTED UPON THE VACUUM CLEANER.

15
HARDWARE PART (CONTD)

• APPROACH 2
• TRANSMITTER RECEIVER IN THIS THE COMPUTER
  GENERATES 4 DIFFERENT SIGNALS BAESD UPON
  ALGORITHM. THESE SIGNALS ARE THEN PASSED UPON TO
  THE TRANSMITTER WHICH SENDS THEM TO THE RECEIVER
  MOUNTED UPON THE VACUUM CLEANER.

16
TRANSMITTER CIRCUIT
17
RESULTS

• THE SMART VACUUM CLEANER CAN BE MADE AWARE OF THE
  WHERE AND WHEN CONTEXT.
• IT WILL ALMOST ELIMINATE THE USER DISTRACTION.
• BY USING PROPER ALGO IT CAN BE MADE TO WORK WITH
  REASONABLE COMPUTING POWER.
• IT CAN BE MADE TO WORK IN ACCORDANCE WITH THE
  SMART ROOM.

18
CONCLUSION FUTUREWORK

• SMART VACUUM CLEANER IS A REALSTIC APPLICATION.
• THE SOFTWARE CAPABLITIES EXIST BUT FURTHER
  DEVELOPMENT IN HARDWARE NEEDED. (PROTOTYPE IS
  BASED UPON A TOY CAR)
• FURTHER WORK NEEDS TO BE DONE TO MAKE THE CLEANER
  AWARE OF OBJECTS IN THE ROOM.