This package contains a mutlipurpose four-wheeled skid-steer drive robot equipped with sensors such as IntelRealSense D435 Depth Camera, LiDAR, Ultrasonics, GPS, IMU, Magnetometer and a pair of cameras.
Obstacle Avoidance with soft turning using pcl processing (passthrough + voxel + RANSAC + euclidean clustering + centroidal analysis).
Detection of slope angle and ditch depth by using a sonar panel.
Dynamic traversal using GPS and IMU to reach desired location along with sonar obstacle avoidance.
Ability of localisation by using two independent EKFs.
Teleoperation with respect to waypoints/time by path smoothening.
Alvar Gate Traversal.
Tennis Ball Detection.
Fully integrated with move_base:
Custom Global Planner:
Traffic Light Model:
You can use the helper script to download all these repositories to the directory above the current path:
./clone_repos.sh
Install all the required ROS related dependencies:
rosdep install --from-paths src --ignore-src -r -y
To install all python related dependencies:
pip install -r requirements.txt
To install all Mapviz related dependencies:
./mapviz_install.sh
For basic bot simulation:
roslaunch atreus xacro.launch
To add an additional python dependency to the list, modify requirements.in
and add a trailing library to the file. After which you need to install the following dependencies in order to use pip-compile
:
pip install pip-tools launchpadlib
Finally use pip-compile
to generate a requirements.txt
file from requirements.in
:
pip-compile requirements.in
Run the following command if the traffic light does not glow red:
./traffic_light_dep_fix.sh
Run the following commands if the map cache in mapviz does not appear.
Based on the errors displayed, correspondingly run the commands and then launch mapviz:
1) ERROR: 1 or ERROR: 203
sudo docker run -p 8080:8080 -d -t -v ~/mapproxy:/mapproxy danielsnider/mapproxy
2) ERROR: 2 or ERROR: 401
sudo docker container ls
sudo docker stop {container id}
sudo docker run -p 8080:8080 -d -t -v ~/mapproxy:/mapproxy danielsnider/mapproxy
3) ERROR: 302
Configured correctly. Poor internet connection.
An experienced AI member of an emerging Rover Development Team. Skilled in ROS, Gazebo, Python, C++, RPi and Arduino. Pursuing BTech in Mechatronics in MIT.
GitHub Repositorygazebo pcl localization obstacle-avoidance teleoperation global-planner lidar artag ball-detection traffic-light imu gps opencv ros robotics