agriopenlink ws@efita 2015

Post on 17-Aug-2015

21 Views

Category:

Food

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

agriOpenLink Key Facts

Interdisciplinary Research Project

- FTW - Research Center for ICT

- JR - Josephinum Research

- BOKU - University of Natural Sciences BOKU

- MKWE - Animal Monitoring System (SMARTBOW)

The main theme: Information Management for Processes Optimization in Precision Agriculture

Funded by: Austrian Research Funding Agency (FFG)

Use Cases: 1) Dairy Farming, 2) Irrigation

Duration: 3 Years ( 06’2013- 05’2016)

Budget: ~900 k Euro (80% funding)

2

Precision Dairy Farming Innovations

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

The Data Integration Problem

Advanced

GUIs &

analytics

Advanced

robotics

Dumping

files

Farmer

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Provide decision

support based on

integrated data

Plugin

agriOpenLink

The Vision

DFO

Integrate any system!

Integrate any data!

Integrate knowledge

3rd Party

Software

API for external APPs

API

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Plugin as a Mediator

6

<subject, predicate, object>

Plugin

Services

Robots

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

agriOpenLink @EuroTier

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Current Work

8

Test Installation:

- Farm Edermaier, 80 ~Cows

- St. Peter in der Au (NÖ)

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Diving into the Ontology

9

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Engineering Ontology

How big is the world that we want to describe? How far do we want to go?

How relevant is this all in our project? Where are the system boundaries?

What tools to use?

Shall we go for OWL at all?

Do we really want to use the reasoner?

How expressive the labels should be?

What language (English, German, …?

What are the best practices in defining names? Who will read them?

What can we reuse? Ontologies? Other knowledge, data models, rules?

Can we do something wrong?

How we will integrate with others?

10

Operational

knowledge

Domain

knowledge

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

agriOpenLink Ontologies

11

Service &

Plugin

Ontology

DF

Ontology

(core)

Exiting Domain

Knowledge but now

put in the format

DF

Ontology

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Service Ontology

12

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Service Description: LelyGetMilkings24h

13

LelyGetMilkings24h

LelyGetMilkings24h

labelcomment

Returns number of milkings.

_node_169

decoratesWith

_node_17_node_195

decoratesWith

decoratesWith

has ServiceNameText

LelyGetMilkings24h

Returns number of milkings.

has ServiceDescriptionText

providedBy

_node_134providedByPlugin

_node_113_node_134

hasOperation

type organization

publisher

FTW

creator

laz@ftw.at

authoritative

false

LelyPlugin

hatLebensNummer

Owl:someValuesfromXsd:string

Xsd:string

hatName

Xsd:int

hatAnzahlMelkvorgaengeProKuh_24h

_node_9

_node_8

operation

paramete

r

typetype

type

Input

objectType

OutputMilkings24h

objectType

OutputParameter

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Domain Ontology – Core (Properties)

14

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

QUERIES

For all cows with the rumination time less than 400 [min/24 h], show

the rumination time and the number of milkings in the last 24 hours.

SELECT ?id ?rumination ?milkingquer

WHERE {

?id livestock:hatAnzahlMelkvorgaengeProKuh_24h ?milkingquery

?id livestock:hatWiederkauaktivitaet_24h ?ruminationquery

FILTER (?rumination < 400)

}

16

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

QUERIES

Get all cows with average milk yield [kg/7 days] higher than 230 [kg]. For each cow show food consumption [kg] and the time spent feeding [s].

SELECT ?id ?grain ?grain_meal ?vitaminE ?milk ?visitquery

{

?id livestock:hatGesamtmengeMilch_7d ?milkquery

?id livestock:hatBesuchDauerSum_7d ?visitquery

?id livestock:hatVerbrauchAlpenkorn_7d ?grainquery

?id livestock:hatVerbrauchSchrot_7d ?grain_mealquery

?id livestock:hatVerbrauchVitE_7d ?vitaminEquery

FILTER (?milk > 230)

}

18

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Domain Ontology – Extensions Many, many more properties ….

19

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Revisiting Engineering Ontology

How to make this all scalable?

What can we reuse?

How to integrate with others?

How to build this as a decentralized innovation?

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.21

Linking Exiting Knowledge

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Benefits

Good format for sharing knowledge

Existing terminology and model could be re-used

Particularly interesting if data from many farms are opened

(with restricted access) based on the same model

Next Steps

- SPARQL End Point

- Editor for the Ontology

- Tools, tools, tools,

- Link to the ADED File in OWL (2003) is available !

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

EFITA/WCCA/CIGR 2015, Poznan, Poland, June 29 to July 2, 2015.

Contact

Dr. Slobodanka Dana Kathrin TomicSenior Researcher | FTW | www.ftw.atForschungszentrum Telekommunikation Wien GmbHDonau-City-Straße 1/3 | A-1220 Vienna | Austria+43/1/5052830 -54 | fax -99 | +43/6769129023

top related