py_drone_toast

Python LANDRS Toast Ontology Api Sparql Testbed designed to enable testing of scenarios on the evolving LANDRS drone ontology/knowledge graphs and provide a test platform for drone interfaces.

Notes:

Documentation

1. The config, data_acquisition, graph and ttl folders have their own README files which you can see on Github by navigating to the folder.
2. There is a wiki associated with this repository that provides instructions and insights into the project at https://github.com/landrs-toolkit/landrs-drone-TOAST/wiki

Getting started

1. Getting the Docker container version running with the Ardupilot SITL (Software In The Loop) can be found at https://github.com/landrs-toolkit/landrs-drone-TOAST/wiki/Docker-deployment-for-TOAST-and-SITL. We forked the SITL repository and our version can be found here https://github.com/landrs-toolkit/docker4sitl.
2. Notes on setting up the Python code natively on a Raspberry Pi is here https://github.com/landrs-toolkit/landrs-drone-TOAST/wiki/Raspberry-Pi.
3. There is an associated repository for the working version of the Ontology at https://github.com/landrs-toolkit/landrsOntTest. We are currently using the config branch.

TOAST information

Using the simulator

The program uses a configuration file (py_drone.ini) to select parameters such as the uuid of the drone, source of the turtle files to load initially etc. Typical values for uuid and .ttl location,

[DRONE]
drone_uuid = MjlmNmVmZTAtNGU1OS00N2I4LWI3MzYtODZkMDQ0MTRiNzcxCg==

[GRAPH]
file = ../landrsOntTest/

where ../landrsOntTest/ represents the current LANDRS test KG at https://github.com/landrs-toolkit/landrsOntTest. To run,

python3 py_drone_toast.py

this will parse the graph for for UAV (drone) MjlmNmVmZTAtNGU1OS00N2I4LWI3MzYtODZkMDQ0MTRiNzcxCg== and setup an api on port 5000. It will also load the ttl files from ../landrsOntTest.

Accessing /, /api or /api/v1

http://localhost:5000/

returns

{
    "openapi": "3.0.0",
    "info": {
        "title": "Priscila's Drone API",
        "description": "Python drone simulation for Knowledge Graph testing.",
        "version": "0.0.1"
    },
    "servers": {
        "url": "http://localhost:5000/api/v1",
        "description": "Flask API running on drone."
    },
    "paths": {
        "/sensors": {
            "get": {
                "summary": "Returns a list of sensors.",
                "description": "Sensors hosted on flight controller board.",
                "responses": {
                    "200": {
                        "description": "A JSON array of sensor ids",
                        "content": {
                            "application/json": {
                                "schema": {
                                    "type": "array",
                                    "items": {
                                        "type": "string"
                                    }
                                }
                            }
                        }
                    }
                }
            }
        }
    },
    "basePath": "/api/v1",
    "http://schema.landrs.org/FlightControllerBoard": "http://ld.landrs.org/id/Mjc2MzRlZWUtZGRiYS00ZjE5LThjMDMtZDBmNDFjNmQzMTY0Cg==",
    "http://www.w3.org/1999/02/22-rdf-syntax-ns#type": [
        "http://www.w3.org/2000/01/rdf-schema#Resource",
        "http://schema.org/Thing",
        "http://www.w3.org/ns/sosa/system",
        "http://schema.landrs.org/schema/FlightControllerBoard",
        "http://schema.landrs.org/schema/Product",
        "http://schema.landrs.org/schema/Thing"
    ],
    "http://www.w3.org/ns/sosa/hosts": [
        "http://ld.landrs.org/id/MmUwNzU4ZDctOTcxZS00N2JhLWIwNGEtNWU4NzAyMzY1YWUwCg==",
        "http://ld.landrs.org/id/Y2U1YThiZTYtZTljMC00ZWY3LTlmMzItZGZhZDk4MTJkNDExCg==",
        "http://ld.landrs.org/id/OTNkMDI1YTctZGE4Ny00Y2IyLWI3MzgtMTU2YzVmMDU1MDI4Cg=="
    ],
    "http://www.wikidata.org/prop/direct/P31": "http://www.wikidata.org/entity/Q220858",
    "http://www.w3.org/ns/sosa/Sensor": [
        "MmUwNzU4ZDctOTcxZS00N2JhLWIwNGEtNWU4NzAyMzY1YWUwCg==",
        "Y2U1YThiZTYtZTljMC00ZWY3LTlmMzItZGZhZDk4MTJkNDExCg=="
    ]
}

which lists the 2 sensors. To look directly at the sensor list,

http://localhost:5000/api/v1/sensors

yeilds,

{
    "sensors": [
        "MmUwNzU4ZDctOTcxZS00N2JhLWIwNGEtNWU4NzAyMzY1YWUwCg==",
        "Y2U1YThiZTYtZTljMC00ZWY3LTlmMzItZGZhZDk4MTJkNDExCg=="
    ]
}

To view a sensor,

http://localhost:5000/api/v1/sensors/MmUwNzU4ZDctOTcxZS00N2JhLWIwNGEtNWU4NzAyMzY1YWUwCg==

returns,

{
    "http://www.w3.org/1999/02/22-rdf-syntax-ns#type": "http://www.w3.org/ns/sosa/Sensor",
    "http://www.w3.org/2000/01/rdf-schema#label": "iSentek Three-axis Magnetometer",
    "http://www.w3.org/2000/01/rdf-schema#comment": "iSentek’s three-axis magnetometers use anisotropic magneto resistance (AMR)technology. They offer high spatial resolution, high precision and low power consumption performance. ",
    "http://schema.org/sameAs": "http://www.isentek.com/en/the_product.php?pid=4",
    "http://www.wikidata.org/prop/direct/P31": "http://www.wikidata.org/entity/Q333921",
    "http://www.w3.org/ns/sosa/isHostedBy": "http://ld.landrs.org/id/Mjc2MzRlZWUtZGRiYS00ZjE5LThjMDMtZDBmNDFjNmQzMTY0Cg==",
    "http://www.w3.org/ns/sosa/observableProperty": "http://sweetontology.net/propSpaceLocation/Position"
}

Available endpoints

• /api/v1 get this drone information and some (now outdated) openAPI information.
• /api/v1/graph get a list of graphs in turtle format.
• /api/v1/graph/ get a list of graphs in turtle format, <api/v1/graph/nFmUsVasTtKGOcNJzhAIDw> a rdfg:Graph;rdflib:storage [a rdflib:Store;rdfs:label 'SQLAlchemy']..
• /api/v1/graph/nFmUsVasTtKGOcNJzhAIDw get the contents of a graph in turtle format.
• /api/v1/id/uuid retrive information on a uuid.
• /api/v1/mavlink start/stop MavLink communications with action=start/action=stop.
• /api/v1/sensors get a list of sensor uuids.
• /api/v1/sensors/uuid retrive information on a sensor by uuid.
• /api/v1/sparql The spaqrql query endpoint. Allows insert, construct as well as query.
• /api/v1/store/OBSERVATIONCOLLECTION/OBSERVATION> save data to OBSERVATION in OBSERVATIONCOLLECTION. * creates OBSERVATIONCOLLECTION. Typical data {"type": "co2", "co2": "342", "time_stamp": "2020-07-11T15:25:10.106776"}.
• /api/v1/turtle/FILENAME download a turtle file of the entire graph to FILENAME.
• /id The id endpoint exposes the URIs for objects created on the drone.
• /sparql The drone hosts a yasgui SPARQL editor webpage here, pointed to the /api/v1/sparql endpoint. Allows insert as well as query.

Other information

1. Uses SQLite database via SQLAlchemy.

Docker container

To build, docker build --tag toast:1.0 ..

To run, docker run --publish 8000:5000 -dit --name toaster toast:1.0. Note the container runs the API on port 5000 but when we run it we moved it to 8000.

To kill, docker rm --force toaster.

Tests

To run tests python3 -m unittest tests/test_graph.py from module root.

Currently test for graph load and storage test.

Planned enhancements,

• generate a hydra http://www.hydra-cg.com compliant api, hydra branch.
• emulate flights to generate data.
• Add endpoint to set/get drone configuration.

Completed enhancements,

• MavLink GPS data capture and storage.

Versions,

1. py_drone_toast.py, sparql queries using rdflib and local base.ttl file
2. py_drone_toast_ld_landrs.py, sparql queries on ld.landrs.org