Fastslam1.0_ROS

This repository is based on EKFSLAM_FastSLAM and Turtlebot3
It is required to install these two packages to obtain the full process
The purpose of making this project is to understand the basic knowledge of SLAM
This is the performance video LINK

Installation

1. Install Turtlebot3
   There are lots of resource teaching people how to install Turtlebot3.
   Here is one of the example, LINK
   Based on the Ubuntu version to choose the right version.
2. Install EKFSLAM_FastSLAM
   Go to the website above,

   git clone <HTTPS or SSH>

   It is suggested to clone this package under the turtlebot3_slam/src file
   Make sure both packages install properly
3. Install this repository
   Clone the package also under the turtlebot3_slam/src file
   It would be easier to modify

Implementation

1. Replace the file
   Copy all the files in this repocitory and replace the same name file under EKFSLAM_FastSLAM
2. Launch Turtlebot3

   roslaunch turtlebot3_gazebo turtlebot3_world.launch

3. Move the Turtlebot
   Open a terminal under this repository and run

   python move.py

   The turtlebot should wait for a few second and move to the certain position
   The initial position of turtlebot is (-2, 0), and the goal position is (0, 2)
   Certain position can be easier change through goal in move.py
4. Checking the collected data
   After Running the move.py, odom_data.dat which includes sensor data would occur.
   Plugging the numbers into the sensor_data.dat
   The data would be ready to perform fastslam
5. Performance
   Open a terminal under EKFSLAM_FastSLAM file and run

   python Main.py

   This will show 2 image, one for EKF filter, another one for fastslam1.0
   Open another terminal under EKFSLAM_FastSLAM file and run

   python TruePath.py

   This will show 1 image of the true path that turtlebot passed through
   The result would occur in few seconds.

Result

TruePath

EKFSlam

FastSLAM1.0

FastSLAM1.0 with particles

Environment

Ubuntu 20.04 noetic
Python 3.6