Branch Name | Languages | Aim | Assumptions |
---|---|---|---|
1. soft-test | Python | testing algorithms | complete info: postition, borders, no transfer lag |
2. hard-test | Python, C++, ROS | other programs are 'safe' | algorithms that are not being tested will be set to 'safe' outputs |
3. master | Python, C++, ROS | final testing | no assumed information |
The simulation of the tag-planner is straight forward to run. The CVXOPT Python library must be installed. The simulation can then be run from the root.py executable. When run the user will have two options for the output, and three for the type of environment being implemented.
- Threshold Plots:
- options: polygon-world (p), sphere-world (s) or all-world (a)
- Animation:
- options: polygon-world (p), sphere-world (s) or all-world (a)
The MATLAB directory contains the Simulink and init.m files which launches the Manager FSM. To launch...
- Launch MATLAB and Simulink.
- Open the FSM diagram titled game.slx.
- Before running the FSM, initialize the game variables by running the init.m script.
- The robots must have position data posted to the amcl_pose ROSTOPIC posted before the FSM can be run correctly.
- The FSM can now run.
The terminal commands that list the outputs of the FSM are:
- rostopic echo bernard_pose
- rostopic echo scrappy_pose
- rostopic echo oswaldo_pose
This folder communicates directly to the roscore during game run-time. Assuming the roscore and rospkg is initialized properly, and the robots are localizing to the same /map topic, the tag planners can be launched using the following command.
rosrun tag planner.py RobotName __name:=RobotName
Where RobotName is the unique identifier for the robot being launched. Currently, the robots are strictly named bernard, scrappy and oswaldo. All robots that will be playing tag should be launched individually.