The efarms project: a two-way road from small farms into the networked world

Transcription

1 The efarms project: a two-way road from small farms into the networked world Claudia Bauzer Medeiros Institute of Computing, UNICAMP (Nov 2007 May 2010)

2 The efarms project: a two-way road from small farms into the networked world Claudia Bauzer Medeiros Institute of Computing, UNICAMP (Nov 2007 and continuing)

3 Outline Motivation and Goals Results Lessons learnt and ongoing work

4

5 DATA

6 MODELS

7 COOPERATION

8 The 2-way road COOPERATION

9 GOALS Science & Technology Sensor-based agricultural data management Models, methods, algorithms Data management, integration, analysis, visualization Low cost wireless data communications in rural areas => Infrastructure

10 GOALS Social Impact Decision support in agriculture Models and software tools Wireless data communications in rural areas => Low cost digital inclusion

11 Expertise involved UNICAMP Computer Science (Databases) (Image Processing) (Compilers and computer architecture) (Wireless networks) UNICAMP Agriculture Engineering (Agricultural model development) (Soil Sciences) (Remote Sensing) One (very) large coffee cooperative (COOXUPE) 14,000 farms (Fourth biggest in the world)

12 Cooxupé exports

13 SOME CHALLENGES Technological Sensor network deployment Scientific Data integration

14 GEO Grupo de Estudos em Geoprocessamento Scale changes 90 t/ha 55 t/ha 75 t/ha

15 RESULTS

16 Results overview Publications 6 journal papers (Geoinformatica 2010, Information Sciences 2010) 15 conference papers (IEEE escience 2010, ACM GIR 2010, ACM GIS 2009, VISAPP 2009) 1 poster, 1 short paper Submitted 2 journal, 1 conf Technological results Low cost networks Prototypes (software and hardware) Crop yield models Training in multidisciplinary research Financed by project - 4 undergrad, 1 MSc Involved Agriculture 3 PhD, 1 MSc Involved CS 4PhD, 2 MSc, 6 undergrad

17 Results (Low cost) Networks

18 Phase 1 Lab Tests

19 Phase 4 - Sensor 3 m pole

20 More sensors

21 MOVIE (SBT TV)

22 Daylight Jan 22, 2010 Underground Humidity (2 levels) Daylight Temperature (air, soil)

23 Daylight Jan 22, 2010 Underground Humidity (2 levels) Daylight Temperature (air, soil)???

24 Daylight Jan 22, 2010 Underground Humidity (2 levels) Daylight Temperature (air, soil) Two-way road

25 Image Processing Efficient algorithms for semiautomatic crop identification

26 COFFEE IDENTIFICATION (Genetic programming)

27 Time series mining Series co-evolution

28 Input 2 nd Closest 3 rd closest Farthest Input: Data from multiple sensors

29 Input 2 nd Closest 3 rd closest Farthest

30 Annotations Workflow driven semantic annotations

31 Arabica coffee, Monte Santo Get_ Annotatio n Schema Get Provenan ce Data County Classify, segment Obtain Yield Valida - <metadata> - <extendinfo> - <location> <county> <state> <country> </location> - <product> <name> </product> </extendinfo> </metadata> 31

32 Interoperability (Web services + Software components)

33

34 Data from local server Data published from Azure

35 Main efforts Sensor callibration and soil experiments Models for yield forecasting Image processing (with Genetic Programming) Databases and interoperability in escience Time series mining Content annotation Sensor data management Escience infrastructure and interoperability

36 Ongoing work and Lessons learnt

37 TODAY Real Farms & the 2-way road

38 Some lessons Multidisciplinary projects need long-term cooperation Sensor response depends on crop Deployment of sensor networks need LOTS of expertise in telecommunications and electronics sensor maintenance Real users need real systems, not prototypes Implanting technological changes is a long process

39 We convinced COOXUPE administrative board and engineers

40 We convinced COOXUPE administrative board and engineers But they need to convince the farmers!!!!!

41 Monitoring Sensor Networks?

42 Monitoring Meteorological Stations Real time data publication Sensors Temperature Air humidity Rainfall Solar radiation Wind speed and direction Event monitoring

43

44 Series mining Annotations Interoperability Complex event processing

45 THANK YOU