- Ashish Belokar (Computer Engineering)
- Kshitij Dholakia (Computer Engineering)
- Swapnil Kate (Computer Engineering)
Implementation of slam_gmapping
The project involves the development of a mobile robotic platform, equipped with wheel encoders and Kinect sensor and using it to carry out Localization and Mapping in indoor environments.
Key implementation points in the project are:
-
Developing mobile robot
-
Converting encoder data to pose estimates (x, y, theta)
-
Identifying Coordinate frames of the robot, and implementing Coordinate frame Transformations
-
Set up the Kinect to stream Laser Scan data from it's depth camera
-
Perform slam_gmapping
-
Visualize the entire process
-
Robot Operating System
-
Kinect
-
Arduino
-
Custom-made Robot: + Differential (rear-wheel) drive with a front castor-wheel + Laptop-mountable Chasis
-
Install Robot Operating System on Ubuntu 11.10 ROS Electric
sudo apt-get install ros-electric-desktop-full
-
Install OpenNI, the Kinect driver, as a ROS Stack OpenNI
sudo apt-get install ros-electric-openni-kinect
-
Download the Arduino IDE, for programming the Arduino microcontroller Arduino IDE
sudo apt-get install arduino
-
Set up Arduino to run as a ROS node using rosserial rosserial
hg clone https://kforge.ros.org/rosserial/hg
-
Set up pointcloud_to_laserscan ROS package pointcloud_to_laserscan
hg clone https://kforge.ros.org/turtlebot/turtlebot
-
SLAM References (Analytical)
- OpenSLAM, a collection of open source SLAM algorithms
- SLAM for dummies, an introduction to SLAM