Data backends / storages are used to store extensional data representing statements that are true. KnowRob does not have a default backend, but rather requires the user to configure which backend to use. The main motivation for this is that reasoner might require specific types of backends, and thus multiple storages must be supported and potentially synchronized with each other.

Additional backend types can be implemented via plugins that implement a common interface which is either the DataBackend or QueryableStorage class. The DataBackend class provides basic functionality for storing and retrieving data, while the QueryableStorage class provides additional functionality for querying the stored data. Currently, KnowRob supports data backend implementations in the C++ and Python languages.

The following queryable backends are available as builtin in KnowRob:

Redland
MongoDB
Prolog:rdf_db

Which storages are initialized is determined through configuration parameters. An example for the MongoDB backend is shown below:

"data-backends": [
  {
    "type": "MongoDB",
    "name": "mongodb",
    "host": "localhost",
    "port": 27017,
    "db": "knowrob"
  }
]

Above, "type" refers to a builtin storage type. If, in addition, "library" is specified, the backend is loaded from a shared library. Alternatively, "module" can be used to load a storage from a Python module. The "name" field is used to refer to the storage in the configuration and also within the runtime. The other fields are specific to the storage type.

Doxygen subpages