- Nader Ahmed
- Winter 2024
This repository consists of several ROS packages
nuturtle_description: visualizes multiple turtlebots of different colors in rvizturtlelib: provides SE(2) transformation methods, SVG visualization, and EKF SLAMnusim: initializes simulation environment with walls and obstaclesnuturtle_control: controls the turtlebot and updates its odometry estimate (integration tests included)nuslam: applies the EKF SLAM to estimate the position of the robot and map the environment
To run the simulator and control with teleop, execute:
ros2 launch nuturtle_control start_robot.launch.xml robot:=nusim cmd_src:=teleop use_rviz:=true
To run with physical hardware, ssh into the turtlebot and run:
ros2 launch nuturtle_control start_robot.launch.xml robot:=localhost cmd_src:=none use_rviz:=false
On your own laptop, run:
ros2 launch nuturtle_control start_robot.launch.xml robot:=none cmd_src:=none use_rviz:=true
The odometry estimate after returning to the initial position was x:.44507, y:0.0002199.
To move the robot in a circle, cmd_src can be changed to circle and the following service must be called with a velocity and radius:
ros2 service call /control nuturtle_control/srv/Control {"velocity: 0.2, radius: 0.5"}
final.mp4
To run the SLAM launchfile and visualize all three robots in rviz:
ros2 launch nuslam slam.launch.xml robot:=nusim cmd_src:=circle
The following service call then moves the simulated turtlebot (red) move in a circle with a velocity of 0.2 m/s and radius of 1.0 m:
ros2 service call /control nuturtle_control/srv/Control {"velocity: 0.2, radius: 1.0"}
The blue robot and its path represents its odometry estimate, and the green robot and its path represents the EKF SLAM estimate.
As seen, the odometry estimate is a perfect circle (as it is unaware of noise and wheel slippage in real world conditions).
The estimated path from SLAM is seen by the green landmarks.
