design and implementation of an agricultural monitoring...

18
Design and Implementation of an Agricultural Monitoring System for Smart Farming Jan Bauer, Nils Aschenbruck University of Osnabrück Institute of Computer Science Distributed Systems Group

Upload: hatruc

Post on 11-Jan-2019

221 views

Category:

Documents


0 download

TRANSCRIPT

Design and Implementation of an Agricultural Monitoring System for Smart Farming

Jan Bauer, Nils Aschenbruck University of Osnabrück

Institute of Computer Science Distributed Systems Group

http://sys.cs.uos.de/ 4

© John Deere Precision Farming • crop yield optimization • precise management, incl.

o fertilization o irrigation o identification of crop diseases

Demand for situational awareness • How to determine and monitor

crop parameters?

Motivation: Agriculture is changing…

© Claas

© Claas

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 6

Leaf Area Index (LAI) • indicator for photosynthetic performance and yield-limiting processes • key variable for models in agronomy, climatology, ecology, …

Motivation: Bio-Physical and Bio-Chemical Crop Parameters

LAI = green leaf area

ground surface area

fractional cover

biomass fPAR

leaf chlorophyll content

OSNBR NDVI

LAI

Crop Parameters

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 7

Basic Concept: Smart fLAIr in a Nutshell

Methodology • radiation-based LAI estimation • Ambient Light Sensor (ALS) • above (A) and below (B) canopy light measurements LAI derived from light transmittance (Monsi and Saeki, 1953)

𝐿𝐿𝐿 = −𝜔 ∗ 𝑙𝑙𝐵��̅�

LAI = 0 LAI = 1 LAI = 3 LAI = 0 LAI = 1 LAI = 3

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 8 © 2018 Institut für Informatik, Universität Osnabrück

Sens4Crop

Jan Bauer, Bastian Siegmann, Thomas Jarmer, Nils Aschenbruck "On the Potential of Wireless Sensor Networks for the In-Situ Assessment of Crop Leaf Area Index" Elsevier Computers and Electronics in Agriculture, Vol. 128, Oct. 2016, pp. 149–159.

http://sys.cs.uos.de/ 9 © 2018 Institut für Informatik, Universität Osnabrück

Smart fLAIr

Jan Bauer, Bastian Siegmann, Thomas Jarmer, Nils Aschenbruck "Smart fLAIr: a Smartphone Application for Fast LAI Retrieval using Ambient Light Sensors" Proc. of the IEEE Sensors Application Symposium, SAS, Catania, Italy, April 20-22, 2016.

Idea • above + below light measurements

via smartphone • Ambient Light Sensor (ALS)

optional: sensor accessory, (i.e., diffuser and band-pass filter)

Prototype • Samsung Galaxy Nexus (GT-i9250) • Android 4.4.4 Kitkat (not rooted!)

Lars Hunning, Jan Bauer, Nils Aschenbruck "A Privacy Preserving Mobile Crowdsensing Architecture for a Smart Farming Application" accepted for the ACM Workshop on Mobile Crowdsensing Systems and Applications in conjunction with the 15th ACM Conference on Embedded Networked Sensor Systems, SenSys, Delft, The Netherlands, Nov. 5-8, 2017.

http://sys.cs.uos.de/ 10

Sensor data transport & IoT Integration

Farm /

FMIS sender receiver

Internet

802.15.4

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 11

Architectural Overview

© 2018 Institut für Informatik, Universität Osnabrück

MQTT Broker

Farm / FMIS

IoT Gateway

User

Cluster n

G G

G G

RPi

Inte

rnet

PL

NM

W

LAN

W

SN

Cluster n+1

G G

G G

RPi R R

http://sys.cs.uos.de/ 12

User Interface

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 13

Architectural Overview

© 2018 Institut für Informatik, Universität Osnabrück

Imag

ery

© G

oogl

e

http://sys.cs.uos.de/ 14

Realization: Technical Aspects

Distributed setup WSN Application • Platform: TelosB • PAR Sensor: Hamamatsu S1087 • Software: TinyOS

Connectivity • Raspberry Pi as Wifi Gateway • Alix UMTS Router for Internet Access

Data Collection • PAR (+ temperature, humidity) • measurement every 2 minutes

• sampling rate: 3 Hz • 10 samples

• centralized LAI processing

Abov

e (A

) sen

sors

Below (B) sensor

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 16

Impressions of harsh outdoor impacts on WSN equipment

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 17

Impressions of harsh outdoor impacts on WSN equipment – Shit happens

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 18

Exemplary time series - Temperature and humidity of two days (single sensor)

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 19

Exemplary time series PAR measurements of a Cluster

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 20

Exemplary time series – LQI values captured by a cluster head

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 22

Conclusion

© 2018 Institut für Informatik, Universität Osnabrück

http://sys.cs.uos.de/ 23

University of Osnabrück School of Mathematics/Computer Science

Distributed Systems Group

Wachsbleiche 27 D-49090 Osnabrück, Germany Phone: +49 541 969 2396 [email protected] http://sys.cs.uos.de

Prof. Dr. Nils Aschenbruck

Thanks for your attention!

?

© 2018 Institut für Informatik, Universität Osnabrück