Beginner
Using Analytical OData in Leonardo IoT
2019-07-14
30 min.
Explore the Analytical OData services in SAP Leonardo IoT (AE) to get time series and aggregates on data
You will learn
- How to use the SAP API Business Hub to browse IoT AE services
- How to use Analytical OData Services from SAP Leonardo IoT
Step 1: Get access to trial systems and tools
Step 2: Access the AE IoT Services API
In this example, we are using data from the enviroCar database https://envirocar.org.
The database contains real-life car data in order to investigate the impact driving has on environmental factors, such as fuel consumption, CO2, or noise emissions. The data is anonymised, meaning it is not possible to connect it to a specific driver. Data available via the enviroCar API can be used by third parties in accordance with the Open Database License (ODbL).
We do not use the enviroCar service directly but a copy and we have further anonymised cars and car types in order to not advertise certain brands.
We will use a part of this data for three cars and a subset of measurements. The three cars will be represented in the package enviroCar of our trial system.
We would like to know, which car had the highest fuel consumption for a given set of tracks. We will obtain the data by using the aggregates function.
In order to verify the thing model, please go to the SAP API Business Hub and search for the API by clicking on Browse:
Enter IOTAE in the search field and press Enter.
The results list will provide the API Package SAP Internet of Things (IoT) Application Enablement services.
Click the package to navigate to the following page:
Choose the tab Artifacts to see the three endpoints to access Configuration, Thing and Thing Analytics. We will have a look at all of them one at a time.
Step 3: Verify configuration
In this step, we will have a look at the configuration service to explore the contents of the package used in this example. We will see the thing types and property set types that we can get while accessing this package. The configuration is similar to a definition of objects and its properties.
Click Configuration to check the package enviroCar. In the following screen, click Get for Configuration {id}.
Scroll down to the editable field for id and enter sap.iotaehandson2.envirocar.
Click Try it out! to make the request.
The response should look like this, showing the package configuration for envirocar with a thing type called cartype and a property set type Measurements.
This is one way to obtain the configuration. You can use a REST client or a similar method to get the same result.
Copy the URL
https://sandbox.api.sap.com:443/iotae/conf/Configuration(%27sap.iotaehandson2.envirocar%27)
and paste it into a REST client like Postman.
Obtain the API key from the upper right corner.
Your API key will show up instead of the gray box. Copy the key to your clipboard.
Add the header apikey with your key.
Click Send to retrieve the following response:
For the following steps, you can use the SAP API Business Hub page to send the request or use a REST client like Postman.
You can request a user for the handson2 tenant which is used in this tutorial and then you can directly call the API like this from a browser or from Postman (with Postman Interceptor). If you want to do so, go to https://www.sap.com/cmp/ft/crm-xm17-gtm-1sc-iotae/index.html and follow the instructions on the page. Afterwards, you are able to access the package configuration using the following link:
https://sap-iotaehandson2.iot-sap.cfapps.eu10.hana.ondemand.com/appcore-conf/Configuration(%27sap.iotaehandson2.envirocar%27)
Step 4: Verify things
Go back to SAP API Business Hub page to select things instead of configuration. Click get for /things.
In case you try this service without a filter, different things from other packages might show up that you do not expect. Please use the following filter to obtain the things as used in this example.
In the edit field for $filter, add the following filter for the thing type:
_thingType eq 'sap.iotaehandson2.envirocar:CarType'
Click Try it out!. This will result in the following response.
You could have chosen the following URL directly:
https://sandbox.api.sap.com:443/iotae/mds/Things?%24filter=_thingType%20eq%20%27sap.iotaehandson2.envirocar%3ACarType%27
The result shows 4 cars that were taken out as examples from the enviroCar database.
Step 5: Access Thing Analytics
Now access the SAP API Business Hub and chose Thing Analytics’. We will verify time series data first.
Click GET for /{propertySetTypeName}/measurements.
Set the following:
-
propertySetTypeName:sap.iotaehandson2.envirocar:Measurements -
$filter:time ge datetime'2018-01-01T00:00:00' and time lt datetime'2018-03-01T00:00:00'
These are a mandatory parameters for the request.
The response will look like this:
The response contains single measurements that were obtained from a car and sent to the enviroCar database. Each single measurement contains the car ID (corresponds to the thing name), a timestamp and a list of different measurements for speed, temperature, fuel consumption, CO2 and more.
The list of measurements looks like this:
"__metadata": {
"id": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/measurements(id='B66347EB30EF46D8BFCD6361532000A5',time=datetime'2018-01-11T06%3A20%3A27')",
"uri": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/measurements(id='B66347EB30EF46D8BFCD6361532000A5',time=datetime'2018-01-11T06%3A20%3A27')",
"type": "sap.iotaehandson2.envirocar.Measurements"
},
"id": "B66347EB30EF46D8BFCD6361532000A5",
"time": "/Date(1515651627000)/",
"Rpm": null,
"Throttle_Position": null,
"GPS_HDOP": null,
"MAF": null,
"GPS_PDOP": null,
"Intake_Pressure": null,
"Intake_Temperature": "14.0",
"O2_Lambda_Voltage_ER": null,
"GPS_Speed": null,
"GPS_Bearing": null,
"GPS_VDOP": null,
"Speed": "5.467811",
"Consumption": "0.7384494",
"CO2": "1.735356",
"GPS_Altitude": null,
"Engine_Load": "36.03984",
"GPS_Accuracy": null,
"O2_Lambda_Voltage": null,
"Track": "5a5709d844ea85087e6841de",
"thingType": "sap.iotaehandson2.envirocar:CarType",
"path": "Measurements"
},
{
"__metadata": {
"id": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/measurements(id='B66347EB30EF46D8BFCD6361532000A5',time=datetime'2018-01-11T06%3A20%3A21')",
"uri": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/measurements(id='B66347EB30EF46D8BFCD6361532000A5',time=datetime'2018-01-11T06%3A20%3A21')",
"type": "sap.iotaehandson2.envirocar.Measurements"
},
"id": "B66347EB30EF46D8BFCD6361532000A5",
"time": "/Date(1515651621000)/",
"Rpm": null,
"Throttle_Position": null,
"GPS_HDOP": null,
"MAF": null,
"GPS_PDOP": null,
"Intake_Pressure": null,
"Intake_Temperature": "13.0",
"O2_Lambda_Voltage_ER": null,
"GPS_Speed": null,
"GPS_Bearing": null,
"GPS_VDOP": null,
"Speed": "4.0",
"Consumption": "0.7563482",
"CO2": "1.7774183",
"GPS_Altitude": null,
"Engine_Load": "37.110767",
"GPS_Accuracy": null,
"O2_Lambda_Voltage": null,
"Track": "5a5709d844ea85087e6841de",
"thingType": "sap.iotaehandson2.envirocar:CarType",
"path": "Measurements"
}
To combine single measurements to a trip, a dedicated Track is added to the measurement (not during driving but later in the database).
Now open the chapter Time Series Aggregates and click GET for /{propertySetTypeName}/aggregates.
Set the following:
-
propertySetTypeName:sap.iotaehandson2.envirocar:Measurements -
$filter:time ge datetime'2018-01-01T00:00:00' and time lt datetime'2018-03-01T00:00:00'
These are a mandatory parameters for the request.
Click Try it out! to obtain the following response:
For each property, a list of characteristics is calculated and displayed (like for CO2):
"CO2_MIN": "1.163272",
"CO2_MAX": "46.601913",
"CO2_AVG": "11.82093400057939",
"CO2_SUM": "69542.55472540855",
"CO2_COUNT": "5883",
"CO2_STDDEV": "8.721946886525672",
"CO2_FIRST": "14.333569",
"CO2_LAST": "1.735356",
"CO2_TFIRST": "/Date(1514541690000)/",
"CO2_TLAST": "/Date(1515651627000)/",
"CO2_TMIN": "/Date(1515159505000)/",
"CO2_TMAX": "/Date(1515563111000)/",
It is possible to filter for a specific car (or thing). Set the following:
-
propertySetTypeName:sap.iotaehandson2.envirocar:Measurements -
$filter:id eq 'B66347EB30EF46D8BFCD6361532000A5' and time ge datetime'2018-01-01T00:00:00' and time lt datetime'2018-03-01T00:00:00'
This will provide the results for a single car only.
The result will look like this:
{
"d": {
"results": [
{
"__metadata": {
"id": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/aggregates('time%3D2018-01-01T00%3A00%3A00Z%2Cduration%3DALL%28T1416H%29%2Cid%3DB66347EB30EF46D8BFCD6361532000A5%2CthingType%3Dsap.iotaehandson2.envirocar%3ACarType%2Cpath%3DMeasurements')",
"uri": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/aggregates('time%3D2018-01-01T00%3A00%3A00Z%2Cduration%3DALL%28T1416H%29%2Cid%3DB66347EB30EF46D8BFCD6361532000A5%2CthingType%3Dsap.iotaehandson2.envirocar%3ACarType%2Cpath%3DMeasurements')",
"type": "sap.iotaehandson2.envirocar.Measurements_aggregates"
},
"_row_id": "time=2018-01-01T00:00:00Z,duration=ALL(T1416H),id=B66347EB30EF46D8BFCD6361532000A5,thingType=sap.iotaehandson2.envirocar:CarType,path=Measurements",
"id": "B66347EB30EF46D8BFCD6361532000A5",
"time": "/Date(1514764800000)/",
"duration": "ALL(T1416H)",
"thingType": "sap.iotaehandson2.envirocar:CarType",
"path": "Measurements",
"GPS_Altitude_MIN": null,
"GPS_Altitude_MAX": null,
"GPS_Altitude_AVG": null,
"GPS_Altitude_SUM": null,
"GPS_Altitude_COUNT": null,
"GPS_Altitude_STDDEV": null,
"GPS_Altitude_FIRST": null,
"GPS_Altitude_LAST": null,
"GPS_Altitude_TFIRST": null,
"GPS_Altitude_TLAST": null,
"GPS_Altitude_TMIN": null,
"GPS_Altitude_TMAX": null,
"GPS_Altitude_PERCENT_GOOD": null,
"Track_MIN": "5a57086244ea85087e67fec2",
"Track_MAX": "5a5709d844ea85087e6841de",
"Track_COUNT": "5883",
"Track_FIRST": "5a57086244ea85087e67fec2",
"Track_LAST": "5a5709d844ea85087e6841de",
"Track_TFIRST": "/Date(1514541690000)/",
"Track_TLAST": "/Date(1515651627000)/",
"Track_TMIN": "/Date(1514541690000)/",
"Track_TMAX": "/Date(1515650476000)/",
"Track_PERCENT_GOOD": "100.0",
"O2_Lambda_Voltage_MIN": null,
"O2_Lambda_Voltage_MAX": null,
"O2_Lambda_Voltage_AVG": null,
"O2_Lambda_Voltage_SUM": null,
"O2_Lambda_Voltage_COUNT": null,
"O2_Lambda_Voltage_STDDEV": null,
"O2_Lambda_Voltage_FIRST": null,
"O2_Lambda_Voltage_LAST": null,
"O2_Lambda_Voltage_TFIRST": null,
"O2_Lambda_Voltage_TLAST": null,
"O2_Lambda_Voltage_TMIN": null,
"O2_Lambda_Voltage_TMAX": null,
"O2_Lambda_Voltage_PERCENT_GOOD": null,
"Consumption_MIN": "0.4950094",
"Consumption_MAX": "19.8306",
"Consumption_AVG": "5.030184682318695",
"Consumption_SUM": "29592.576486080885",
"Consumption_COUNT": "5883",
"Consumption_STDDEV": "3.7114667588897077",
"Consumption_FIRST": "6.099391",
"Consumption_LAST": "0.7384494",
"Consumption_TFIRST": "/Date(1514541690000)/",
"Consumption_TLAST": "/Date(1515651627000)/",
"Consumption_TMIN": "/Date(1515159505000)/",
"Consumption_TMAX": "/Date(1515563111000)/",
"Consumption_PERCENT_GOOD": "100.0",
...
"Speed_MIN": "0.0",
"Speed_MAX": "139.0",
"Speed_AVG": "72.54980761738969",
"Speed_SUM": "426810.5182131035",
"Speed_COUNT": "5883",
"Speed_STDDEV": "43.86616899863244",
"Speed_FIRST": "27.934315",
"Speed_LAST": "5.467811",
"Speed_TFIRST": "/Date(1514541690000)/",
"Speed_TLAST": "/Date(1515651627000)/",
"Speed_TMIN": "/Date(1514547393000)/",
"Speed_TMAX": "/Date(1515131392000)/",
"Speed_PERCENT_GOOD": "100.0",
"GPS_Speed_MIN": null,
"GPS_Speed_MAX": null,
"GPS_Speed_AVG": null,
"GPS_Speed_SUM": null,
"GPS_Speed_COUNT": null,
"GPS_Speed_STDDEV": null,
"GPS_Speed_FIRST": null,
"GPS_Speed_LAST": null,
"GPS_Speed_TFIRST": null,
"GPS_Speed_TLAST": null,
"GPS_Speed_TMIN": null,
"GPS_Speed_TMAX": null,
"GPS_Speed_PERCENT_GOOD": null,
"GPS_Accuracy_MIN": null,
"GPS_Accuracy_MAX": null,
"GPS_Accuracy_AVG": null,
"GPS_Accuracy_SUM": null,
"GPS_Accuracy_COUNT": null,
"GPS_Accuracy_STDDEV": null,
"GPS_Accuracy_FIRST": null,
"GPS_Accuracy_LAST": null,
"GPS_Accuracy_TFIRST": null,
"GPS_Accuracy_TLAST": null,
"GPS_Accuracy_TMIN": null,
"GPS_Accuracy_TMAX": null,
"GPS_Accuracy_PERCENT_GOOD": null
}
]
}
}
Step 6: Access Thing Analytics with further options
If you have processed Step 5, we can now consider additional options or verify details by enhancing the request with further parameters.
Remain in the open chapter Time Series Aggregates and click GET for /{propertySetTypeName}/aggregates.
Set the following:
-
propertySetTypeName:sap.iotaehandson2.envirocar:Measurements2 -
$filter:time ge datetime'2018-03-01T00:00:00' and time lt datetime'2018-05-01T00:00:00'
These are a mandatory parameters for the request.
$select should be as follows:
Track,Speed_AVG,Consumption_AVG
Click Try it out! to obtain the following response:
{
"d": {
"results": [
{
"__metadata": {
"id": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/aggregates('time%3D2018-03-01T00%3A00%3A00Z%2Cduration%3DALL%28T1464H%29%2CTrack%3D5a4f5d0144ea85087e3b5d9e')",
"uri": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/aggregates('time%3D2018-03-01T00%3A00%3A00Z%2Cduration%3DALL%28T1464H%29%2CTrack%3D5a4f5d0144ea85087e3b5d9e')",
"type": "sap.iotaehandson2.envirocar.Measurements2_aggregates"
},
"Track": "5a4f5d0144ea85087e3b5d9e",
"Consumption_AVG": "7.097110271453857",
"Speed_AVG": "32.0"
},
{
"__metadata": {
"id": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/aggregates('time%3D2018-03-01T00%3A00%3A00Z%2Cduration%3DALL%28T1464H%29%2CTrack%3D5a53ba3044ea85087e433cba')",
"uri": "https://analytics-thing-sap.cfapps.eu10.hana.ondemand.com:443/aggregates('time%3D2018-03-01T00%3A00%3A00Z%2Cduration%3DALL%28T1464H%29%2CTrack%3D5a53ba3044ea85087e433cba')",
"type": "sap.iotaehandson2.envirocar.Measurements2_aggregates"
},
"Track": "5a53ba3044ea85087e433cba",
"Consumption_AVG": "0.9329440295696259",
"Speed_AVG": "1.5499999523162842"
}
...
]
}
}
Compared to the example in Step 5, the aggregates are calculated based on a further dimension: the track number. This was possible as the property set Measurements2 contained a different setting for property Track. In Measurements2, it was marked as a dimension.
So you see that you can slice and dice the data by using Analytical OData services. If you compare the data for all obtained results, you will find that the highest consumption average occurred in the last track 5a5709d844ea85087e6841de as the average speed was the highest, too.
In case that you want to know which car or thing was used in the track, please add id at the beginning of the select list.
For further questions regarding the API, please refer to the latest documentation: