Category Archives: Sensing

Scouting by Consumer UAV

Consumer UAVs are increasingly seen as farm tools.  Some come with camera and packaged tech for easy flying, pretty much straight out of the box.

But before you leap in, please be aware there are RULES.

We suggest you spend time on the AirShare www.airshare.co.nz and CAA www.caa.govt.nz/rpas/ websites before you get started.  Designed specifically for UAV users they have easy to digest information setting out what you can and cannot do.

DJI Phantom 3

Our package came with all equipment, an extra battery and optional propeller guards packed in a tough custom carry case.  The camera is on a gimbal for steady shots, panning and tilting. Zoom in by getting closer!

A downloaded smartphone or tablet app shows flight information such as height, position and battery charge and lets you see exactly what the camera sees with no delay.

In windy conditions, we achieved about 13 minutes of flight time rather than the 23 minutes stated for each battery charge. Rules say you must be able to see the aircraft with your own eyes so you are probably limited to under 100ha. You could make a reasonable inspection in that time.

Peas and onions from 30m Web

We used the UAV to scout at the LandWISE MicroFarm. Viewed from 30m up, crop variation is immediately obvious.  Pea flowering striping seems to match drill widths. We had variable emergence too so ponder the link. Sprayer runs are visible too.

On the onion side we see thinner areas to the bottom right, and patches where Plant & Food have harvested sample plants as part of our joint OnionsNZ research project.

OnionsFrom40m_VerticalPlots

Viewed from directly overhead we see more of Plant & Food’s research plots, some harvested and some still being followed through to final harvest. The image indicates all these plots are within a reasonably good and even part of the crop.

To the bottom right, a lower wetter area shows lower populations where plants are smaller and fewer made it through establishment.

OnionsFromUAV_CloseUp45_web

Dropping to a metre of two above the crop and tilting the camera, we see up close. Because we are seeing what the camera is seeing, we can choose exactly what we want to check and go there immediately.

So we’ve scouted the whole paddock, had a closer look here and there, and if we need to, we can walk to the spots we want to check in detail. The thing is, we know where we should be looking.

Satellite Imagery

A large part of Heretaunga Plains horticulture was photographed for us by satellite at the end of November.

World View 2 satellite coverage of the Heretaunga Plains on 23 November 2105
World View 2 satellite coverage of the Heretaunga Plains on 23 November 2105

Part of our OnionsNZ Variability project, the World View 2 coverage targeted our crop and other onion crops east of Hastings.

By capturing four bands of light, Blue, Green, Red and Near Infrared, we are able to get a “normal” colour image like an aerial photo, and a biomass map using the NDVI index.

The satellite image pixel size in 0.5m x 0.5m, so we get at least two pixels across each onion bed.

World View 2 NDVI image captured 23 november 2015 of MicroFarm onion and vining pea crops
World View 2 NDVI image captured 23 November 2015 of MicroFarm onion and vining pea crops

In the NDVI image, the onion crop is lower left paddock, the vining peas upper right. Red areas indicate low or no biomass, yellow light, green moderate and blue heavy cover. Note however that the value of each colour is slightly different for each crop.

Because the pea canopy is full ground cover while the onions are only roughly half ground cover, we had to use different value bands to see variation within each crop. If we used the same range, either the peas would all be blue, or the onions mostly yellow and red.

The striping effect in the onions is the onion beds. Some adjacent beds have quite different canopy densities.  The red edge around the onions is bare soil and light canopy in the outer beds. The blue area in the centre is influenced by charcoal from an old bonfire site. Even taking these things into account, there is a reasonably large amount of variation in this crop.

Red spots in the pea crop are patches with no plants. The red headlands show light canopy areas and the red strip centre right the irrigator access track. There are three different seed lines of Ashton peas making up the pea crop. These are not discernable in the satellite image. The crop was harvested on 14 December, and there was no significant difference seen in hand harvested plots or in the viner.

Onion variability Year 1

OnionsNZ

Enhancing the profitability and value of New Zealand onions

The purpose of this OnionsNZ MPI Sustainable Farming Fund research project is to provide the industry with tools to monitor and manage low yields and variability in onion yield and bulb quality.

In this collaboration with Plant & Food Research, LandWISE is providing precision agriculture paddock scale measurement and interpretation.

We have base maps from topography and surface ponding analysis completed by Page Bloomer Associates, and from AgriOptics Dual EM soil mapping. We also have some previous crop data including true colour, false colour and NDVI images of winter cover crops between successive onion crops in these paddocks. More details here>

We tracked crop development with a range of sensor technologies including AltusUAS MicaSense from UAV, Agricultural Software GroundCover app and some satellite imagery.

The collaboration with Plant & Food Research was to help us develop protocols to monitor crop development and yield variation (spatially and temporally). Linking these with crop modelling and agronomy helps determine why variation is occurring.

Crops were traced from paddock through harvest and storage so that post-harvest quality issues can be related to factors during growth. Linking paddock production to packhouse performance and back again may be key in unlocking value potential.

Grower led focus groups are involved in the project and analysis of results. They have a vital role in the development of practical tools they can use to monitor and quantify variability, to identify the causes of loss of yield and quality and share best practice to improve sustainability and grower returns.

 

Onion Research Underway

OnionsNZ

After months of planning our OnionsNZ, Plant and Food, Sustainable Farming Fund onion variability project, things are underway at the MicroFarm!

Gerry Steenkamer planted the crop on 2nd August. Rhinestone seed was donated by Vigour Seeds and treated for us by Seed and Field Services. We are very grateful for their support.

An initial residual herbicide application of Dacthal and Stomp was applied. We had a lot of wireweed last year and are keen to get on top of that.

PlantandFoodwebPlant and Food Research staff have established plots for detailed monitoring. They are doing many very detailed individual plant measurements at plot scale. LandWISE is coordinating a number of sensing surveys of the whole crop using a range of technologies.

20150805_PlotMarking
Plant and Food staff setting up monitoring plots in onions beds. (Wintery southerly)

More details for the research programme and measurements are available on the MicroFarm website.

Plant and Food researchers have developed growth models for a range of crops. This work will help refine their onion growth model, a key to understanding the development and variability in crops. The detailed plot measurements will also be compared with the whole paddock sensor measurements to corroborate and calibrate them.

The first paddock scale surveys have been completed. These give some base information and understanding of the site and it’s variability. Maps as pdfs are available on the MicroFarm website.

One of the first “layers” we can look at is Google Earth imagery – free info on the web! Have a look at your place: use the time slider to view a series of aerial and satellite images captured over recent and not so recent years.

 

MicroFarm Onion Beds with Winter Cover Crops (as shown on Google Earth image 19 April 2015)
MicroFarm Onion Beds with Winter Cover Crops (as shown on Google Earth image 19 April 2015)

We have posted some of “our place” images and some interpretation here>.

Pagebloomer vsPage Bloomer Associates completed an RTK-GPS survey using Trimble equipment from GPS Control Systems. The data were used to create surface ponding and runoff risk maps.

agriopticsAgriOptics completed a Dual-EM survey in early July. This gave shallow and deep soil information maps. The dry winter means soil had not reached field capacity when the survey was made, so we are a little cautious when interpreting the results. But we risked not getting a survey at all, and by planting in August it had still not rained.  With beds formed and crop planted and emerging, we have no opportunity now to repeat the survey.

AltusUAVAltusUAS has prepared NDVI maps of cover crops from UAV mounted sensors. They will be making repeated measurements as the crop develops. AltusUAS is now using MicaSense technology for efficient multispectral image collection.

 

ASL_Square_150ASL Software has provided their Cover Map canopy cover measurement tool fitted with high accuracy GPS. We can now use that technology to measure relative plant development and ensure our readings (our mapped data) are located in the correct beds!

BioRich Tractor for MicroFarm

The arrival of a BioRich sponsored tractor at the LandWISE MicroFarm will support precision farming research efforts.

BioRich_JD900HCWeb
The John Deere high clearance cropping tractor is set to match the onion beds at the MicroFarm.

The John Deere cropping tractor has been set to a 183cm wheel track to fit the onion beds planted in early August. It’s first role is to act as a carrier of sensors that are used to map crop development.

We are delighted with the tractor. After much investigation into options for a sensor carrier, we finally landed on a high clearance cropping tractor as the ideal machine. Then, after searching wide and long, we discovered there was one sitting on our back door.

BioRich Principal, Mike Glazebrook is a LandWISE founding member and past Chairman. He said he was keen to support  the work being done at the MicroFarm as he sees it as of benefit to the community. There is obvious alignment with LandWISE objectives for sustainable production.

BioRich Limited is an organic waste recycling company. It’s main activity is capturing organic material that would otherwise be wasted, or cause pollution, and turning it into rich compost. Where it is practical to do so it also seeks to extract stock food and energy from organic material that would otherwise be wasted.

Every year, throughout New Zealand, many thousands of tonnes of organic “waste” is dumped into landfills or is inappropriately discharged to land. Once dumped much of this material breaks down in an uncontrolled manner and releases greenhouse gases into the atmosphere and pollutants into our waterways.

Meanwhile most of New Zealand’s cultivated soils have been steadily deteriorating. This is due to both to a decline in soil organic matter and a depletion of minerals and nutrients.

Hence BioRich’s mission is to divert organic matter (carbon) from ending up in places where it can do a lot of harm – in our atmosphere and water – and putting it somewhere it can do a lot of good – in our soils.

Pioneering Precision

Aerial Imaging for Better Data Collection

Rural Delivery Series Eleven, Episode 24
Saturday 22 August 7am TVOne

The use of sensors of one kind or another is nothing new in agriculture, particularly in the hands of cropping farmers.  But a new imaging tool is currently being evaluated at Massey University.  It was originally designed for space exploration and military operations but is now being adapted for data collection, to help farmers make the best management decisions possible.

The sensor is flown over land, gathering images from more than 450 wavebands including visible, near, short and infra-red.  Maps of farms are developed, identifying pasture quality, nutrient content, potassium and sulphur levels, land surface temperatures, and areas of poor drainage as well as nutrient movements on slopes.

Professor Ian Yule of Precision Agriculture at Massey University says in the past, remote sensing has tended to focus on nitrogen use but the more sophisticated sensors allow the presence and concentration of other nutrients to be determined.

The technology is being used as part of Pioneering to Precision, a Primary Growth Partnership Programme (PGP) with particular interest in fertiliser application on hill country, but Professor Yule says it will also have significance for the dairy industry.

Onions Research – three year project

LandWISE has partnered with Onions New Zealand and Plant & Food Research in a three year project focused on understanding variability in onion crops. The project is funding by Onions NZ and the MPI Sustainable Farming Fund.

Dr Jane Adams, OnionsNZ Research and Innovation Manager, says the project, “Enhancing the profitability and value of New Zealand onions” is designed to provide the industry with tools to monitor and manage low yields and variability in onion yield and bulb quality.

It will incorporate precision agriculture with initial work to be done at the LandWISE MicroFarm. At the MicroFarm, we have been building increasing knowledge of the site, but will ramp that up with more layers of soil and crop information as we try to unpick factors contributing to lower yields and reduced quality.

Information about the 2014-2015 MicroFarm Onion crop can be found on the MicroFarm website.

The project proper starts on 1 July, but there has a lot of preparatory activity to ensure everything kicks of smoothly.

Anyone interested in joining a regional Focus Group supporting the project should
contact us>

OnionsNZ

 

LandWISE 2015: The Farm of 2030

20-21 May 2015, Havelock North, NZ

Farmof2030Web

Looking for the next big thing in agriculture?

Maybe the next big thing is a small thing. Bigger equipment has given amazing work efficiencies and helped drive production and productivity.  Are there any downsides?

As we seek to achieve efficient crop production, we need to manage variability. What are the drivers, what tools can help us? How do we link technology and agronomy?

Leave LandWISE 2015 with new understanding of where technology has taken us, where current development is opening possibilities and which things may yet be some way off.

The theme “The Farm of 2030” comes from a prediction made in 1980 by John Matthews of the UK National Institute for Agricultural Engineering Describing a farm fifty years in the future, soil quality and alternative machinery featured strongly. Computers and robotics were becoming available but GPS, internet and wireless were not.

NIAE_Cover

Join us at our 13th Annual Conference, a meeting of technologists, farmers and their many support providers, where you can engage with leading researchers and practitioners.

Top presenters from Australia and New Zealand will update you on sensors, networks and robots, crop and field variability and what we can do to manage it. Dealing with troublesome weeds, identifying pest outbreaks, monitoring soil moisture and automating irrigation management. Can a robot do our scouting, our weed control, our mowing?

Ultimately, what will a farm look like in 2030? What do we need to be doing to make sure we are ready?

Register for LandWISE 2015 and be part of the “Design-a-Bot Challenge”!

More here>

Sponsor_Sheet_600

The Mechanical Farm of 2030

LandWISE 2015 Presenter, Dan Bloomer

 

DanBloomer200
Dan is the Manager of LandWISE Inc, an independent consultant, and a member of the Precision Agriculture Association of New Zealand Executive.

In 1981, John Matthews of the UK National Institute for Agricultural Engineering described what a farm would look like in 2030; a fifty year horizon.

“The mechanical farm of 2030” identified four factors that would influence the farm of 2030; social factors including employment, preservation of the environment, animal welfare and primary energy sources.

NIAE_Cover

Soil quality and alternative machinery were high on their list. Computers and robotics were available but GPS, internet and wireless were not.

In 2015, with all the benefits of knowing what happened in the last 35 years, we revisited the question to ask, “What will a cropping farm look like in 2030?” Were John Matthew’s predictions of technology on-track? And importantly, what must farmers do to ready themselves for next year, five years and fifteen years down the track?

The general consensus was a resounding round of applause for John Matthews. The issues he identified continue to be key drivers today. The technological developments he envisioned are progressing towards the 2030 deadline with examples of commercially developed gantries now being tested on farms in Europe.

MatthewsGantry
The NIAE Gantry image from John Matthews paper
The ASALift Gantry tractor in 2013
The ASALift Gantry tractor in 2013

John Matthews article included a robotic harvester. We know the computing and actuation required for that is still tricky, but it seems quite probable robotic harvesting will be feasible and possible it will be relatively common by 2030.

The NIAE robotic harvester image from Matthews' paper
The NIAE robotic harvester image from Matthews’ paper

Perhaps his control tower windows are more likely to be computer monitors, and he didn’t know about smart phones, but his vision of the role computing would play is remarkably close – though perhaps thanks to Moore’s law and compounding development we have already got further than he estimated.

MatthewsComputer
The NIAE image of a farm computer appears to have a rack for storage disks, but also shows a microphone and aerial perhaps for wireless communications.

Maybe the design (how) is different to now, but much of the what of John Matthews’ predictions suggests he deserves a high score.

 

Agri-Intelligent Systems: robots, data, and decisions

LandWISE 2015 Presenter, Tristan Perez

Tristan Perez Professor of Robotics and Autonomous Systems,  Queensland University of Technology, QLD, Australia
Tristan Perez
Professor of Robotics and Autonomous Systems,
Queensland University of Technology, QLD, Australia

Since the 1960s, agriculture has seen significant advances in agrochemicals, crop and animal genetics, agricultural mechanisation and improved management practices. These technologies have been at the core of increased productivity and will continue to provide future incremental improvements. Data analytics, robotics, and autonomous systems are transforming industries such as mining, manufacturing, and health. We are starting to see automation of single agricultural processes such as animal and crop remote monitoring, robotic weed management, irrigation, nutrient decision support, etc. However, we envisage that the integration of these technologies together with a systems view of the farming enterprise and its place within the agri-food value chain will trigger the next wave of productive innovation in agriculture.

The challenge of the next agricultural revolution is to assist farming enterprises to make the management and business decisions that will optimise inputs such as labour, energy, water and agrochemicals and explicitly account for variability and uncertainty across the production system and along the agri-food value chain.

The opportunity for increased profitability, sustainability and competitiveness from finer-scale sensing and whole-farm decision-making and intervention requires farmers to have greater access to digital data and technologies to extract information from data. The agricultural landscape will rapidly change due to low-cost and portable ICT infrastructure.

Agri-intelligence is the integrated collection of tools and techniques – from robots, unmanned airborne vehicles (UAVs) and sensor networks to sophisticated mathematical models and algorithms – that can help farmers make sense of large amounts of data (agronomic, environmental and economic) to make risk-informed decisions and run their farms more profitably and sustainably.

Perez AgriIntelligence

 

The figure below shows the ubiquitous emerging vision of a farm in the second machine age, where computer systems are used to augment human perception and capacity for decision making in complex situations.

PerezAgriIntelligentFarmSystem

The farming enterprise is considered a system that interacts with the environment (through climate, markets, value chain, etc.) The key objective is to make sound decisions about management in order to optimise inputs, yield, quality, and at the same time make the system robust against yield and quality volatility due to climate, commodity market fluctuations, and incomplete information about the state of crop, soil, weeds and pests.