Welcome to the Attention System repository. This system is designed to manage a list of attention points, moving the robot's head from one point to another while keeping a record of how often each point has been visited. This ensures a dynamic interaction environment, focusing on areas of interest based on predefined criteria or external inputs.
- Git: Required for cloning the repository. If not installed, download it from https://git-scm.com/.
- ROS2 humble: Ensure ROS2 humble is installed on your system to run the attention system. For installation details, refer to the ROS2 Installation Guide.
Create a new workspace or use an existing one. To create a new workspace, run the following commands in a terminal:
mkdir -p ~/attention_system_ws/src
cd ~/attention_system_ws/srcClone the attention system repository with the following command:
git clone https://github.com/aaggj/attention_systemBefore running the system, ensure to compile the workspace with ROS2. Navigate to your workspace's root directory and execute:
cd ~/attention_system_ws
colcon build --symlink-install
source install/setup.bashTo run the attention system, use the following ROS2 command:
ros2 run attention_system attention_serverThis command will start a lifecycle cascade node that will be configured and activated automatically, listening on the /attention/attention_points topic. This topic expects messages of type attention_system_msgs/msg/AttentionPoints, which are used to manage the attention points within the system.
The AttentionPoints message is designed to contain specific information about the attention points. Be sure to review the message definition in the repository to understand the structure and required fields.
If you are interested in contributing to the project, we encourage you to do so. You can start by reviewing the open issues in the repository, discussing new features, or reporting bugs. For major contributions, please make sure to open a new issue to discuss your ideas before making a pull request.