‘IoT-ready’ Sensor Technology for
Connected Farms
http://scampp.com/
Krikor Ozanyan and Bruce Grieve
School of Electrical and Electronic Engineering
<[email protected]>; <[email protected]. UK>
Sensors in 1st generation IoT
First wave: ‘All Data Are Sensed’
• Data acquisition
COTS sensors for basic measurands• miniaturisation
• low consumption circuits
• energy harvesting
• wireless-connected
• Data processing
Basic sensor fusion ‘tripod’• Architectures
• Algorithms
• Applications A
AA
http://www.gambica.org.uk/resourceLibrary/-the-internet-of-things-tree-of-life.html
Livestock farms - the supply chain
Food
Farm
Abattoir
Managing the chain -
essential for efficiency and resilience
• Upstream (e.g. food)• ingredients
• processing and storage
• formulation
• transport
• Downstream (e.g. abattoir)• transport
• schedule
• numbers/weight
Connected Farms in context: Improve food conversion
Food intake is accountable by volume/weight
Animal mass can be monitored by weight
• Food intake is regular• Animal weighing is much more challenging to manage
regularly and accurately• Existing weighing technology is either expensive and/or
o requires substantial maintenance; o requires staff engagement; o depends on manual/visual handling; o results in low throughput;o delivers raw data onlyo is difficult to connect
https://www.ritchie-d.co.uk
C‘OTS’ methodology needed
• Inexpensive, multiple weighing sites (100% covered)
• Allows regular weighing, as part of the farm routine
• Easy integration of animal ID (tags or camera)
• Unobtrusive for the animals
• Fully connected, upgradeable, re-programmable
• Low farm-floor maintenance, minimal disruption
?
Weighing - mature and widely spread, well commercialised For connected pig farms we need (at minimum overhead)
Borrowing iMAGiMAT™ technology
Translation to livestock farms: • Deployment in hostile environments – materials ‘sandwich’ to allow
reliable encapsulation without compromising sensitivity• Connectivity and flexibility (upgrading/re-programming)• Several application scenarios, based on sensing floor deformation
•New tomography type sensing, originally developed for human gait
•Based on deformation of plastic optical fibres, scales well in area
•Materials cost: ~£100 m-2
IEEE Sensors J. 15 (1) 279-289 (2015)
doi: 10.1109/JSEN.2014.2341455 .
Borrowing iMAGiMAT™ technology
Data acquisition• iMAGiMAT is based on collaborative sensor fusion from multiple POF sensors
arranged as a carefully designed grid in a single plane • No wires and no electrical currents under the mat, electronics at the periphery
Data processing• Imaging is by solving an inverse problem – special methods developed for
reconstruction at severe undersampling e.g. Hough transform.• Real-time reconstruction is speeded up by parallel computation in embedded
multiprocessor systems / FPGAs• Automatic decision making without visualisation – classifications by artificial
intelligence (deep neural networks)• Resources and communications efficiency is achieved by balancing “on-board”
and “hub” processing, as well as balancing real-time vs post-processing.
www.manchester.ac.uk/eee/e-agri
Agri-Food IoT, e.g. Above Crop Below Soil and beyond the Farm Gate
Sensors for protecting crop YIELD
● A sensor-network of 24/7 in-field disease pressure monitors
● Sensors “trick” pathogen into germination on mimic surface
● £2.5M leveraged funding, TSB Crop Protection competition
Sub-Soil Imaging
● New visualisation tool for seeds
breeders
● Tracks how efficiently the root
bundles are in drawing upon the
water and nutrients in the soil.
● An in-field tool for early isolation
and delivery of tomorrow's climate
tolerant food crops
Plastic Electronic-sensor Tagging (PET)
● UK, 4.1M tonnes of food that could have been
eaten is thrown away
● A technology platform for recording produce
temperature stress profiles from low cost,
battery-free, plastic RFIDs
● £1.5M leveraged funding from TSB Plastic
Electronics competition
N8 AgriFood – www.n8agrifood.ac.uk
One Network, Many Solutions
Multiuse Networked Platforms, e.g.IoT Crop “Tricorder” for < US$100
Connected GPS, camera, microprocessor
+Subtle light manipulation
= Active Close-Proximity Hyperspectral Imaging Fluorescence Imaging Stereo photometric Imaging Solid State Active Ellipsometry
Image courtesy of CBS Inc.
• Crop disease sensing• Early weed / plant detection;• Speciating insect disease-vectors;• Carbon footprint (protein in crops);• Leaf viral symptoms;
• Harvested fruit bruising & sugars• Soil organic carbon• Aquatic parasites• Frogs in salads ...
The near horizon
• Cloud – balance between data and information
• Big Data management, IoT+
• (C)OTS start – ‘bolt-on’ IoT capability necessary
• Standards – for modular structures where applicable
• People – education and training compatible with the spread and evolution of IoT