Cases from RIVM, Weather Station and Smart Emission

Sensor Web Enablement
in practise
Thijs Brentjens & Michel Grothe
TU Delft, 4 juni 2015
Agenda
1. Applications of sensor web (enablement) -MG
2. SWE for Air Quality in The Netherlands - TB
- Intro Pilot project
- SOS operations demo
- Client demos
3. INSPIRE as drive force behing SWE – MG
Break
4. New (IoT) kid on the block; OGC SensorThings API
–
–
–
Internet of Things Intro – MG
SensorThings API –TB
Application smart emissions - TB
5. Living Lab ‘Making Sense for Society’ -MG
6. Final remarks & invitation
Q&A
1. Applications of sensor web enablement
The application of the sensor web enablement
(SWE) standards is still in it’s infancy,
compared to the application of other OGC geo web
standards like CSW/WMS/WMTS/WFS/...
Most SWE standards are matured and tested
thoroughly in many projects all over the world.
1. Applications of sensor web
enablement
Tsunami early warning system South East Asia
GITEWS
1. Applications of sensor web
enablement
Federal Administration of German Waterways
Wupperverband
1. Applications of sensor web
enablement
Air quality monitoring in Europe - EAA
1. Applications of sensor web
enablement
Citizen Eagle watch
1. Applications of sensor web
enablement
1. Applications of sensor web
enablement
1. Applications of sensor web
enablement
French groundwater level monitoring
1. Applications of sensor web
enablement
RITMARE (Italian research for the sea)
SOS
sos:insertObservation
sos:insertObservation
1. Applications of sensor web
enablement
Geonovum weatherstation
http://sensors.geonovum.nl
1. Applications of sensor web
enablement
Sensor web enablement Projects in EU








eo2Heaven
S@NY
COBWEB
GENESIS
OSIRIS
ENVIROFI
Eenvplus
…
1. Applications of sensor web
enablement
Sensor web (enablement) catalog
www.geocens.ca
2. SWE for Air Quality in The
Netherlands
2. SWE for Air Quality in The Netherlands
Pilot project
●
Nationwide data on several types of emission,
measured from several locations
(http://www.lml.rivm.nl/ )
●
http://www.lml.rivm.nl/meetnet/index.php
●
E-reporting air quality European Union
●
Includes zones, all kinds of metadata and real
measurements
2. SWE for Air Quality in The
Netherlands
Pilot project
●
Sensor Observation Service + more
●
Open Source software
●
Source code:
https://github.com/Geonovum/sospilot
●
Docs:
●
http://sensors.geonovum.nl/
●
http://sospilot.readthedocs.org/en/latest/
●
Author: Just van den Broecke
2. SWE for Air Quality in The
Netherlands
Pilot project
2. SWE for Air Quality in The
Netherlands
A brief look at SOS requests
●
Basic Observation Model
–
Procedure
–
FeatureOfInterest
–
Phenomenon
–
Observation
2. SWE for Air Quality in The
Netherlands
A brief look at SOS requests
●
●
●
GetCapabilities
http://sensors.geonovum.nl/sos/se
rvice?request=GetCapabilities&vers
ion=2.0.0&service=SOS
List service operations,
offerings,
featuresofInterest,
observableProperties etc
2. SWE for Air Quality in The
Netherlands
A brief look at SOS requests
●
GetObservation
–
http://sensors.geonovum.nl/sos/service/pox
2. SWE for Air Quality in The
Netherlands
A brief look at SOS requests
<GetObservation
xmlns="http://www.opengis.net/sos/1.0"
version="1.0.0" service="SOS"
xmlns:om="http://www.opengis.net/om/1.0"
xmlns:ogc="http://www.opengis.net/ogc"
xmlns:gml="http://www.opengis.net/gml"
xsi:schemaLocation="http://www.opengis.net/
sos/1.0
http://schemas.opengis.net/sos/1.0.0/sosGet
Observation.xsd"
xmlns:xsi="http://www.w3.org/2001/XMLSchema
-instance">
2. SWE for Air Quality in The
Netherlands
A brief look at SOS requests
●
FeatureOfInterest
–
All:http://sensors.geonovum.nl/sos/service?service=SOS
&version=2.0.0&request=GetFeatureOfInterest
–
–
Or specific feature with ID:
http://sensors.geonovum.nl/rivmlml/procedure/742
http://sensors.geonovum.nl/sos/service?service=SOS&ve
rsion=2.0.0&request=GetFeatureOfInterest&featureOfInte
rest=http://sensors.geonovum.nl/rivmlml/featureOfInterest/742
2. SWE for Air Quality in The
Netherlands
A brief look at SOS requests
●
http://sensors.geonovum.nl/sos/client
●
http://sensors.geonovum.nl/sos-js-app/
2. SWE for Air Quality in The
Netherlands
Disadvantages SOS
●
●
●
Verbose: sometimes lots of data
Generic: maybe too much possibilitites /
flexibility?
XML: lots of web & app developers prefer JSON
over XML
2. SWE for Air Quality in The
Netherlands
Access data through other API's
●
●
“Lightweight API” → JavaSript sensorweb
version / JSON data
–
Non-standardized API, 52North specific
–
Ease of implementation
http://sensors.geonovum.nl/jsclient#map
2. SWE for Air Quality in The
Netherlands
Access data through other API's
●
Regular WMS (WMS-Time) and WFS:
–
Bit limited possibilities for sensordata
–
RIVM case:
●
●
●
useful for showing a demo of stations and
(some) measurements
ease of using other software libraries
http://sensors.geonovum.nl/heronviewer/
2. SWE for Air Quality in The
Netherlands
REST API Sensordata
●
●
●
Sensordata of the RIVM prototype in another
API:
http://sensors.geonovum.nl/sos/api/v1/station
s
http://sensors.geonovum.nl/jsclient/#map
3. INSPIRE as driving force
behind SWE in Europe
3. INSPIRE as driving force behind SWE
3. INSPIRE as driving force
behind SWE in Europe
INSPIRE SWE with O&M and SOS
● INSPIRE as Framework Directive is not only focused on the
’spatial’
o ex : Environmental Monitoring Facilities definition : “...
includes observation and measurement of … by or on behalf
of public authorities.
● Thus: Necessity to provide observation data in Annex II and
Annex III
=> Guidelines for the use of Observations & Measurements and
Sensor Web Enablement-related standards in INSPIRE Annex II
and III data specification development
● SOS is seen as one of the possible candidates for the extension
of TG for INSPIRE Download services (v. 3.0)
3. INSPIRE as driving force
behind SWE in Europe
O&M Data Models in INSPIRE
7 Themes integrating Observations
A. Geology
B. Oceanographic Geographical Features
C. Atmospheric Conditions
D. Environmental Monitoring Facilities
E. Soil
F. Species Distribution
G. Natural Risk Zones
3. INSPIRE as driving force
behind SWE in Europe
O&M Data Models in INSPIRE
Possible future extensions
a. Area management/restriction/regulation zones
...
b. Human Health and Safety
c. Land cover
d. Production and industrial facilities
e. Statistical units & Population
distribution,demography
f. Utility and governmental services
g. Habitats & biotopes
3. INSPIRE as driving force
behind SWE in Europe
SWE Overview
OGC Sensor Web Enablement Suite (SWE) provides base data and
service standards
● Observations & Measurements (O&M):
o Base data model for provision of observational or
measurement data
o Integrated into several INSPIRE data models
● Sensor Observation Service (SOS):
o OGC Webservice for provision of O&M data
o Same structure as other OGC services
o Tailored for access to O&M data with focus on time series
● Sensor Model Language (SensorML)
o Description of measurement process
o In INSPIRE use INSPIRE Process
3. INSPIRE as driving force
behind SWE in Europe
What is an observation?
To understand the data from an observation or measurement, we
must know:
● What was measured (observedProperty)
● Where was it measured (featureOfInterest)
● How was it measured (procedure)
● When was it measured (phenomenonTime)
● Data quality information (resultQuality)
And of course, we need
the result of the observation.
3. INSPIRE as driving force
behind SWE in Europe
SOS in a nutshell
● OGC Standard
● Current version: 2.0
● Applicable when sensor data needs to be
managed in an interoperable way
● Part of SWE suite
3. INSPIRE as driving force
behind SWE in Europe
Sensor Web Enablement suite
3. INSPIRE as driving force
behind SWE in Europe
SOS as an INSPIRE Download Service
● SOS is seen as one of the possible candidates for the extension
of TG for INSPIRE Download services (v. 3.0)
● JRC study on SOS (2014)
o Maturity of clients/servers
o Mapping between SOS 2.0 specs and INSPIRE NS Regulation
o Open Source Implementation (52North SOS)
● MIG Sub-group (MIWP-7a)
3. INSPIRE as driving force
behind SWE in Europe
3. INSPIRE as driving force
behind SWE in Europe
O&M Guidelines
● Provide recommendations and guidance for use of
O&M in INSPIRE
● Design Patterns provide support in structuring
different types of observations
● INSPIRE Extensions are described:
o INSPIRE Process
o Specialized Observations
o Observable Properties
o Options for result encoding
o Referencing Observations
4. New kid on the block; SensorThings API
‘SWE light’
http://ogc-iot.github.io/ogc-iot-api/
4. New kid on the block;
SensorThings API
Internet of Things
4. New kid on the block;
SensorThings API
Internet connected devices
4. New kid on the block;
SensorThings API
SWE is moving towards IoT
4. New kid on the block;
SensorThings API
SensorThings API
●
Based on OGC Sensor Web Enablement
–
Retrieve sensor data
–
Submit tasks to control sensors and
actuators
●
Efficient REST-like API and JSON encoding
●
Support standard-based location encodings
–
indoor/outdoor, mobile/stationary
●
Linked Data ready (JSON-LD)
●
Pub-Sub ready (MQTT)
4. New kid on the block;
SensorThings API
SensorThings API
●
●
Compared to SOS & SPS:
–
Easier to use
–
Less complexity
–
Less functionality
More suitable for small devices, like in IoT
4. New kid on the block;
SensorThings API
SensorThings Data Model
4. New kid on the block;
SensorThings API
SensorThings Data Model
4. New kid on the block;
SensorThings API
SensorThings Data Model
4. New kid on the block;
SensorThings API
{
"id":1,
Location encoding: GeoJSON
"Self-Link":"http://demoURL:8080/SensorThings_V1.0/Locations(1)",
"Things":{
"Association-Link":"Locations(1)/$links/Things",
"Navigation-Link":"Locations(1)/Things"
},
"encodingType":"http://example.com/location_types#GeoJSON",
"location":{
"type":"Point",
"coordinates":[
5.1,
52.15
]
}
}
4. New kid on the block;
SensorThings API
Location encoding: indoor, CityGML
{
"id":2,
"Self-Link":"http://demoURL:8080/SensorThings_V1.0/Locations(2)",
"Things":{
"Association-Link":"Locations(1)/$links/Things",
"Navigation-Link":"Locations(1)/Things"
},
"encodingType":"http://example.com/location_types#cityGML",
"location":{
some CityGML
}
}
4. New kid on the block;
SensorThings API
SensorThings API
●
http://ogc-iot.github.io/ogc-iot-api/api.html
●
REST-Like: HTTP
–
CRUD each resource with HTTP verbs
–
CREATE - HTTP POST
–
READ - HTTP GET
–
UPDATE - HTTP PUT
–
DELETE - HTTP DELETE
4. New kid on the block;
SensorThings API
SensorThings API
●
Example for HTTP GET –-> retrieve data using
simple URLs
http://example.com/SensorThings_V1.0/Data
streams(1)/Observations
●
Returns all the observations in the datastream
●
Compare to GetObservation in XML encoding
●
Response in JSON:
–
Support in all major programming languages
4. New kid on the block;
SensorThings API
SensorThings API
●
Closer to basic web-technology
●
Lightweight: easier for IoT devices
●
Candidate specification: not much (no?) reallife examples available at this moment
4. New kid on the block;
SensorThings API
Smart emissions
Sensor web challenges for cities,
a local air quality monitoring example in the city of Nijmegen
4. New kid on the block;
SensorThings API
Project goals ‘Smart Emissions’
1. Deployment of a local air quality network using a low-cost
sensor network
2. Involvement of citizens in the deployment and
maintenance of the sensor network
3. Involvement of citizens in the analysis of the results of
local air quality monitoring
4. New kid on the block;
SensorThings API
Smart emissions
4. New kid on the block;
SensorThings API
Smart emissions
4. New kid on the block;
SensorThings API
Smart emissions
4. New kid on the block;
SensorThings API
Smart emissions
4. New kid on the block;
SensorThings API
Smart emissions
also sensor web enabled!
4. New kid on the block;
SensorThings API
Issues and (research) questions to deal with
1.
Deployment of a local air quality network using a low-cost sensors








What is the quality of low-cost sensors in general?
Which type of low cost sensors to deploy?
How to calibrate the low-cost sensors?
How many and at what locations (spatial pattern) to deploy the sensors?
What data platform for data collection and distribution?
Which standards for data acquisition and distribution?
Which (interpolation) models for furher processing air quality data?
How to visualise the results?
2.
Involvement of citizens in the deployment and maintenance of the sensor network:
– Which method to use for citizen engagement?
– Do we need to training citizens to deploy and maintain the sensor?
3.
Involvement of citizens in the analysis of the results of local air quality monitoring
– How to visualise the data?
– How to interact maps with citizens?
– How en when to meetup with citizens
– etc.
4. New kid on the block;
SensorThings API
Which type of low cost sensors to deploy?
Quality and price
Number of sensors applied in a city
4. New kid on the block;
SensorThings API
Which type of low cost sensors to deploy?
4. New kid on the block;
SensorThings API
How to calibrate the low-cost sensors?
 Calibration by air quality experts from the
National Institute of Environment and Health
(RIVM)
 Calibration at two national air quality locations in
the City of Nijmegen
4. New kid on the block;
SensorThings API
How many and at what locations (spatial pattern)
to deploy the sensors?

How many?


Goals to achieve?
Where to locate?


Covering the whole City or certain parts of the city (e.g. potential problem areas)
Financial resources (also in case of many low-cost sensors)!
Dilemma: research versus politics
-
Professionals: put sensors at high risk areas for what-if analysis
-
Politicians say: no clustering of sensors in potential high risks areas, because
they rather do not want to specify ‘problem areas’ as such!
4. New kid on the block;
SensorThings API
Data platform for data acquisition?
Question /issues to deal with are related to:
4. New kid on the block;
SensorThings API
Which standards for data distribution?
OGC Sensor Things API
4. New kid on the block;
SensorThings API
Smart emissions
Proposed architecture setup
Applications
Sensorthings
API
Josene
Sensors
WMS
WFS
WCS
SOS
View
service
Download
Service
Download
Service
Download
Service
API
Sensor
observations
Transformation
Service
sensor
data
Download
Service
CityGIS
ETL Service
Sensor
data
O&M
Geonovum
4. New kid on the block;
SensorThings API
The analytics (theory, models and processes)
The interpolation model used:
 Basic GIS interpolation?
 National RIO interpolation model of the
national Institute of Environment?
 Specific air pollution city model?
 Develop an own interpolation model?
But also, further data integration and use
of local air pollution data to e.g.:
-
Citizen engagement
Dynamic traffic management
Human health impact assessment
…
4. New kid on the block;
SensorThings API
The visualisation / communication framework is
strongly related to citizen engagement using
mapping and maptable technique
All results are presented
in maps!
Citizen engagement
organised and supported
by mapping and
maptable
Air quality data will be
open available through
Internet (API) for
further analysis
4. New kid on the block;
SensorThings API
Sensor web for smart cities is just beginning!
 The interest in low-cost sensor networks in cities
is increasing. In the Netherlands approx. 10
larger cities are more or less exploring local air
quality monitoring with low-cost sensor networks.
 We need multidisciplinary experts in these sensor
web city applications with strong citizen
engagement
 There are still several issues to be solved and
research questions to be answered
4. New kid on the block;
SensorThings API
Cases of local air pollution monitoring in the
Netherlands
Smart cities and Smart People
 Approx. 10 cities in NL
 Deployment of local air quality networks
using a low-cost sensors for continous
monitoring
• Through communities and teams with
smart thinkers, makers, developers and
designers, policy makers, …
• Involvement of citizens in the
deployment and maintenance of the
sensor network
• Involvement of citizens in the analysis of
the results of local air quality monitoring
4. New kid on the block;
SensorThings API
Smart emissions
5. Living Lab ‘Making Sense for Society’
“Open platform for the application of
the Internet of Things and People”
Main characteristics
Ambition
Cooperation, knowledge exchange and appointments regarding the application of
the Internet of Things and People.
Open platform
For professionals from private sector, knowlegde institutes and government.
Open access with expert resources on a voluntary basis.
Action oriented
Work towards solutions for society through local cases and cross-cutting themes.
To answer questions, to solve common issues and colloborate for innovations.
Cases and cross-cutting themes
Case 1 –
Citizens and
perception
City of
Zwolle
Case 2 –
Case 3 –
Smart
Emissions
Living Lab
Stratumseind
City of
Nijmegen
City of
Eindhoven
Case 4 –
EnergyThings
campus
Wageningen
University
of
Wageningen
Case 5 –
Case 6 –
Seismic
activities and
Buildings
Day at the
Beach
Companies
Case ? –
….
City of
Den Haag
Policy issues
Privacy & Ethics
Standards
Data science
3D sensor data visualisation
5. Final remarks and invitations
More information:
Pilot AQ & weatherstation Geonovum
http://sensors.geonovum.nl
Living Lab Making Sense for Society
http://www.geonovum.nl/onderwerpen/sensor-geoinformatie/algemeen-living-lab-internet-everything
We invite you to join …
24 juni 5e plenaire meeting
i.s.m. gemeente Nijmegen en
de Radboud Universiteit
Just register at (free access):
http://www.geonovum.nl/onderwerpen/sensorgeo-informatie/nieuws/24-juni-living-lab-internet-everything-bijeenkomst-nijmegen
We invite you to join …
7 oktober 6e plenaire meeting
Sensor Makers Event!
te Nijmegen
www.geonovum.nl/sensor-web-and-internet-things
[email protected] | www.geonovum.nl | @geonovum
Thank you for your attention!
We would like to acknowlegde for their valuable input:
Alexander Kotsev, JRC, ISPRA-Italy
Steve Liang, University of Calgary, Canada