CONVINCE model2code Documentation
The tool is developed as part of the CONVINCE project and is mainly used to generate skill-level code starting from an SCXML model.
The inputs of the program are:
an SCXML model that describes the behavior of the skill,
an XML file that describes the full model of the program, which includes the behavior tree, the skills and the components used in the system,
an XML file that describes the interfaces used between behavior tree and skills, and between skills and components,
a directory containing the templates of the files to generate.
The output of the program is composed of:
a C++ file that contains the code of the skill,
a header file that contains the declaration of the skill,
a main C++ file that contains the main function of the skill,
a package.xml file that contains the information of the ROS package,
a CMakeLists.txt file that includes the generated files.
The output above is only if the data model is ECMAscript, otherwise, you will need to add the parameter --datamodel_mode to generate the C++ data model files.
In this case, you will get 2 additional files:
a C++ file that contains the code of the data model,
a header file that contains the declaration of the data model.
The parameters required to run the program are:
--input_filename: the path to the SCXML file that describes the behavior of the skill,--model_filename: the path to the XML file that describes the full model of the program,--interface_filename: the path to the XML file that describes the interfaces used.
By default the program will generate the code in the directory where the SCXML file passed by the parameter --input_filename is located, but you can specify a different directory by using the parameter --output_path.
The skills generated are based on a behavior tree structure and will have a ROS2 tick service in case they are a condition a ROS2 tick and halt services in case they are an action.