krishna-13-cyber / micromouse-eklavya

MICROMOUSE PROJECT

A micromouse is a small sized, fast bot that can solve a 16x16 in a short period.

Table of Contents

• About the Project
  • Tech Stack
  • File Structure
• Getting Started
  • Prerequisites
  • Installation
  • Execution
• Theory and Approach
• Results and Demo
• Future Work
• Contributors
• Acknowledgements and Resources
• License

About the project

AIM

To make the micromouse bot capable of reaching the centre of the given maze in the shortest possible time using maze solving algorithms. As of now, I have almost implemented the left follow algorithm on it. The right follow algorithm is similar but the opposite. The micromouse bot also somewhat requires PID tuning to stay in the centre of the path.

Tech Stack

• ROS Noetic
• Gazebo 11
• Python 3

File Structure

 ┣ 📂.git
 ┣ 📂assets
 ┃ ┣ 📜Micromouse Problem Statement.pdf                               
 ┃ ┣ 📜README.md
 ┃ ┣ 📜ROS-Learning-resources.md
 ┃ ┣ 📜Screencast from 24-10-21 03 29 56 PM IST.mp4
 ┃ ┣ 📜Screencast from 24-10-21 03 35 50 PM IST.mp4
 ┃ ┣ 📜Screencast from 24-10-21 03 38 05 PM IST.mp4
 ┃ ┣ 📜Screencast from 24-10-21 03 39 31 PM IST.mp4
 ┃ ┣ 📜Screencast from 24-10-21 03 47 10 PM IST.mp4
 ┃ ┣ 📜Screencast_from_24-10-21_03_29_56_PM_IST_SparkVideo.gif
 ┃ ┣ 📜Screencast_from_24-10-21_03_35_50_PM_IST_SparkVideo.gif
 ┃ ┣ 📜Screencast_from_24-10-21_03_38_05_PM_IST_SparkVideo.gif
 ┃ ┣ 📜Screencast_from_24-10-21_03_39_31_PM_IST_SparkVideo.gif
 ┃ ┣ 📜Screencast_from_24-10-21_03_47_10_PM_IST_SparkVideo.gif
 ┃ ┣ 📜Screenshot from 2021-10-24 11-43-34.png
 ┃ ┣ 📜Screenshot from 2021-10-24 11-43-54.png
 ┃ ┣ 📜Screenshot from 2021-10-24 11-44-06.png
 ┃ ┗ 📜Screenshot from 2021-10-24 11-44-55.png
 ┣ 📂config
 ┃ ┗ 📜pos_control.yaml
 ┣ 📂launch
 ┃ ┣ 📜final.launch
 ┃ ┣ 📜gazebo.launch
 ┃ ┗ 📜rviz.launch
 ┣ 📂rviz
 ┃ ┣ 📜rviz_config.rviz
 ┃ ┗ 📜rviz_scan_config.rviz
 ┣ 📂scripts
 ┃ ┣ 📂Algorithms
 ┃ ┃ ┣ 📜bot.py
 ┃ ┃ ┣ 📜node_left_follow.py
 ┃ ┃ ┗ 📜read_laser.py
 ┃ ┗ 📂legacy
 ┃ ┃ ┣ 📜cmd_vel_robot.py
 ┃ ┃ ┣ 📜node_follow_wall.py
 ┃ ┃ ┣ 📜node_go_to_point.py
 ┃ ┃ ┣ 📜node_maze_runner.py
 ┃ ┃ ┣ 📜node_obstacle_avoidance.py
 ┃ ┃ ┣ 📜node_reading_laser.py
 ┃ ┃ ┣ 📜position_controller.py
 ┃ ┃ ┗ 📜velocity_controller.py
 ┣ 📂urdf
 ┃ ┗ 📜micromouse_robot.urdf
 ┣ 📂world
 ┃ ┣ 📜arena.world
 ┃ ┗ 📜my_arena.world
 ┣ 📜.gitignore
 ┣ 📜CMakeLists.txt
 ┣ 📜Micromouse Problem Statement.pdf
 ┣ 📜package.xml
 ┗ 📜readme.md

Getting Started

Prerequisites

1. ROS Noetic
2. Gazebo Version: 11.0
3. RViz (not compulsory)

• You can find the video demonstration link for installtion of ROS Noetic here and the website link for the same here.
• The link to install Gazebo for Ubuntu is provided here. You can find the instructions to install different versions of Gazebo in the link provided above but using Gazebo 11.0 is recommeded for ROS Noetic.
• The link to install RViz on Ubuntu is provided here.

Installation

git clone https://github.com/KalyaniKasar1/Micromouse-Eklavya.git

Add this folder in the src directory of your catkin workspace Create the src folder if it doesn't already exist by

mkdir src

Initialise the project with

catkin_make
source /opt/ros/noetic/setup.bash      #run this and the command below everytime
source ~/catkin_ws/devel/setup.bash     you need to launch nodes 

Execution

Open two terminal windows and run the following commands

• Terminal 1

source /opt/ros/noetic/setup.bash
source ~/catkin_ws/devel/setup.bash
roslaunch pkg_techfest_imc gazebo.launch

• Terminal 2

source /opt/ros/noetic/setup.bash
source ~/catkin_ws/devel/setup.bash
rosrun pkg_techfest_imc node_left_follow.py

Theory and Approach

• I already had the urdf model of the bot required and the model of the maze, which I had spawned in Gazebo. So I didn’t need to design them on my own.
• Starting with the basics, the robot needs to move forward (maximum permitted speed is 0.4 m/s) and take left, right or U turns as per it needs. So for these linear.x and angular.z velocities are published to the cmd_vel topic of the bot in the form of geometry_msgs/Twist messages.
• Next, it needs to see and take in its surroundings in order to perceive what movements it has to make. For this, their is a laser scanner at the front of the bot that sends out 360 beams in an area spanning -90 to +90 degrees in front of it. Each degree of area in front of the bot contains 2 beams. These beams hit the walls of the maze and get reflected back to the bot; they are received by a receiver which can perceive the distance that the reflecting surface for that particular beam was. 360 beams means 360 different values are obtained by the receiver in the form of sensor_msgs/LaserScan messages, which are then published to the /my_mm_robot/laser/scan topic; msg.ranges[ : ] returns a list of values of the distance detected for each beam in the selected range in [ ].
• For determination of rotation, yaw of the bot is obtained from its quaternion values, which in turn are obtained from the /odom topic in the form of Odometry messages. Rotation of the bot is done using the obtained yaw values instead of using time to rotate by 90 or -90 or 180 degrees as required.
• Various algorithms like left or right follow rule, flood-fill, DFS, BFS, Djikstra’s algorithm or A * algorithm, etc can be used. For now, I have implemented the left follow algorithm to solve the maze.
• PID too needs to be used to keep the bot at the centre of the path and so that it doesn’t collide with the walls gradually and make unnecessary turns and increase the time taken to solve the maze. Without PID tuning the bot may also get stuck in a loop of collisions.

Results and Demo

Initial position of micromouse in the maze :

Clips of micromouse when it's finding its way across the maze (using left follow rule):

Screencast.from.24-10-21.03.29.56.PM.IST.mp4

Screencast.from.24-10-21.03.35.50.PM.IST.mp4

Screencast.from.24-10-21.03.38.05.PM.IST.mp4

Screencast.from.24-10-21.03.39.31.PM.IST.mp4

Screencast.from.24-10-21.03.47.10.PM.IST.mp4

Future Work

• Write a code for PID to make sure the bot remains a the centre of the path
• Implement other algorithms
• Work on making it remember the correct turns to be taken at every junction/node so that one final run can be made out of the 7 allowed in which it travels to the centre with the shortest possible path, by storing a text in text file.
• Replace delay everywhere with other conditions
• Make a final.launch such that all the necessary ROS nodes, RViz and Gazebo Simulation to solve the maze can be launched at once.
• Learn about gzmaze so that a maze of any desired structure can be generated

Contributors

• Kalyani Kasar

Acknowledgements and Resources

• SRA Vjti Eklavya 2021
  
• Special thanks to my mentors Gautam Agrawal, Anushree Sabnis and Prathamesh Tagore.
  
• https://github.com/hashmis79/Micromouse

License

The license used for this project.