Simultaneous Localization and Mapping

Under Construction!

This project is still in progress.

What is in this repository?

This repository contains my implementation of FastSLAM algorithm presented in the Probabilistic Robotics book.

In other words, the presented ROS package utilizes the robot's odometry and lidar data to map the environment as well as to estimate the robot's pose by means of Monte-Carlo localization. The developed algorithm is simulated in Gazebo using the ClearPath Husky robot.

Simulation in Gazebo:

Map of the environment (robot location in green):

How to install the package?

Open a terminal and follow the below steps to clone the package

mkdir -p catkin_ws/src
cd catkin_ws
catkin init
cd src
git clone https://github.com/OmarJItani/Simultaneous-Localization-and-Mapping.git
cd ..
catkin_make

How to test the package?

To launch the husky robot in gazebo, open a terminal and type:

export HUSKY_LMS1XX_ENABLED=1;
roslaunch husky_gazebo husky_playpen.launch

To run the slam node, open a new terminal and type:

rosrun fast_slam_husky_gazebo my_husky_fastslam.py

To move the husky robot from your keyboard, open a new terminal and type:

rosrun key_teleop key_teleop.py key_vel:=/husky_velocity_controller/cmd_vel