25th February
After our session yesterday with Lyn in which we were introduced to the year, we were encouraged to talk about our ideas.
While there were many impromptu discussions about our topics, one of the most grounding conversations I had was with Jonathan in which we laid out our ideas and described them to one another to ensure they are clear.
Brainstorming the different terrains: With things that New Zealand and the rest of the world have in common being placed down the centre. The value in this was identifying that New Zealand features a very wide variety of terrain on which agricultural land may exist. I may need to define the type of terrain I am designing for or I may need to ensure that the product I design is able to easily move over the terrain while effectively carrying out its task.
26th February
I had coffee with Daniel from Fablab to discuss the potential of the mechanical parts of my project.
In our discussion he described “The three R’s of robotics”
1.Risky
2. Repetitious
3.Remote
The theory is that if a task is any of two of these three then it warrants an investigation in a robot counterpart to do the job rather than a human. I feel that my project covers all three of these areas as there is risk involved in the current spraying of herbicides and the harmful chemicals that users often come in contact with, it is a repetitious and physical task that takes up the users time and strength. It is often a task that is undertaken on fairly remote land which could be done by a device rather than the farmer.
I found this conversation really motivating and felt really excited about the prospects that the project held in terms of its innovation and benefits that it could offer the user.
______________________________________________________________________
Later that morning, I popped in to speak to Rodney about my topic and to get some feedback on the potential people I could connect with for the project. As if seredipity, Tony Parker also came in and we established contact and he welcomed me to send through any of my work and progress throughout the project, for which I am very grateful.
Rodney also assisted in setting me up with some things to research as an initial springboard, which is much appreciated. I borrowed the following book from Rodney that is full of really interesting contextual information.
5th March
Harry Illife who now works at Fablab came for a walk through fourth year space and we began talking about my project and it was suggested that I look at Farmbot, which is an open-source CNC arm that gets used for maintaining a small garden in residential to life-style block sized gardens.
I found this chat really helpful.
Another observation I made was that the further one gets into your research, the more you start jumping into where you are currently up to rather than beginning from the start for those who have never come into contact with your project.
6th March
Daniel and I spoke about open-sourced Robotics and methods on how I could best inform my design thinking and learning around robotics.
He suggested that watching videos on how things like this are built and how the mechanisms work will be of great benefit to me.
We also discussed methods of powering the device and he suggested that I potentially consider a docking station where the device can both recharge and refill on the mixture.
9th March
At the shared lunch in the shared space studio, I began speaking to third-year Tessa Livingstone around the robotics involved in my project.
She suggested I watch the Netflix documentary series; Love, Death & Robots as it explores some of the ethical issues around the rise of robotics.
In the afternoon session of studio, I had my low-risk ethics documentation signed off, so now I can begin reaching out to different people in the farming community.
12 March
I found some wheels and bits at my flat and decided that building a little frame to try and figure out scale and to have a way of visualising ideas both for myself and to convey these ideas to others would be good.
I went to the workshop (which was nice and empty) where Charlie and I started chatting about my project. She liked the sustainability aspects of my design and we spoke about the potentials in the emerging technology.
Uli then joined the conversation and showed me the steam machines that they have downstairs which I could use to simulate how my design works if it utilises steam to kill the weeds. He also suggested a flame thrower…
We spoke about ways that the device could be charged and he suggested looking at how the electric fences are powered and potentially going there for its recharging.
We also spoke about the wheels and the different variations including having something on independent suspension and how that will affect the overall performance of the device.
I found going to the workshop and speaking to the staff, especially while they are less busy to be incredibly helpful, and will do this more often when I run into snags on the physical modelling of stuff.
20-21 March
Please see the write ups on the Interviews that had taken place 20-21 March which is located in a different tab under the interactions Blog.
5 April
Presenting to the Family
For me, it is really important to continue working on some elements of my projects throughout the lockdown period so that my project can retain some of its momentum even during a very uncertain time. To assist with this, I set up a presentation that was done to my family to show them the progress I have made on my project thus far. This gave me something to work towards.
The following is their feedback:
-Using "Umm" a lot
-Clearer explanation on where the concept comes from
-What is your final aim?
-Explain what dairy farming vs crop spraying needs/ looks like?
-Explain what a suspension system is?
-What is the average size of a paddock?
-Waste minimisation.
-How will you recharge/reload it?
-Does it have a fail safe/ stop button?
-What if it falls over?
-Weather
-Should be more clear at start what you are designing.
-Could you look at other brands more and why people like them?
-Why did you choose to design this?
-What is the benefit of your design?
-It should have a personality.
-Could it be modular and customisable?
-How will it identify weeds? and why this method?
Grandpa: "Lasers..."
I really appreciated the feedback I have received from my family, they did not hold back with the questions and while the university is closed, it is the most constructive feedback I could manage.
Kerry Harrington
Expert- Associate Professor in Weed Science
I recognise that Kerry Harrington is definitely an expert in his field and genuinely appreciate his input into my project. He passed along some readings that should be very beneficial to also aid me in my research exploration and finding out more about some of the specific functions that the device will have to have.
8 April
Wyatt Page
Expert- Associate Professor of Acoustics and Human Health
10am skype meeting with Gus from class and Wyatt.
-He has an image processing background.
-He asked if there could be a team of devices working together.
-Timing the device to work effectively but also fast enough to catch the weeds at a young stage when it does its next cycle of the paddock.
-what quality of imagery needs to be captured to allow the device to make the decision?
-What does the foliage look like at that early stage?
-how the herbicide gets added to the device.
-Look at including soil testing. - Is it valid? Is it overcomplicated?
-COST of current products. - Could you estimate amount of weeds on a paddock.
-The parts that are involved- replacing parts rather than units. - Minimising the waste generated.
-barometer technology
-battery- how to get it to effectively charge
-could be slower for more accuracy.
-Could it have a long distance camera on the front?
-OBSTACLE AVOIDANCE
-Noises- dont want to spook livestock.
-At what scale do you want to eradicate the weeds?- will need to define this.
-Pinpoint suspension
-Service- delivery model. Could it be used by a company that unleashes them in the paddocks or for the farmer directly.
-Multiple of them would mean that the cost would need to be lower so that the user could afford to have more than 1.
-Grass quality visual.
-Could it be sold to contractors?
-More technological research can be done, with a look at more futuristic approaches.
-Case studies of things like the imow to see how they charge.
- Thinking outside the box in terms of concept sketches.
-plug-in sensors.
-What other products can detect p;ants and how can I utilise this?- ie. phones with plant recognition apps.
-App- could it show grass and soil quality
-how to build the interactive display on the device.
-User interface- could be represented by encapsulating a phone.
-There is a real risk of compromising on aesthetic for function in this project.
-Explore different options- explore all avenues and approaches.
-How complex is the data you want from the sensors? affects trade offs.
I found having a chat with Wyatt to be really helpful in terms of generating more questions and areas of exploration for myself. He has also indicated that he would be happy for us to further discuss our projects with him.
Brett Downing
Expert- Former Senior Mechatronic Engineer CSIRO High Resolution Plant Phenomics Centre
4pm meeting via Skype
-Hyper-spectral cameras - pick up colour and fluorescence- with a camera like this you can see a lot of things pertaining to the difference between different plants and this would also be able to indicate some of the qualities of the grass and soil also.
-A regular GoPro or DSLR camera should be able to effectively also take screenshots or use frames from video to identify plant life that it might see as interesting and then use the plant identification technology located in the "brain" to decipher whether it is a weed or not and act accordingly.
-The idea of human error being eradicated is hard, because computer error at a prototyping stage will be high too.
-His work was around crops and learning about methods to test their health.
-plants often exposed to blue light and then there is measurement of how much infra-red light bounces back from them.
-Optical Spot Spraying
-It is hard to update the programmed decision making feature in a "brain"
-Better and more eyes (cameras) can allow for more information to be collected.
-There is software available that is able to tell you what it is doing, this could be beneficial if I have a screen on the device that allows the user to see what phase the device is in.
-Cost effective and reliable parts will be key.
-THE CAMERA NEEDS A DRIVER.
-How many sensors you need will have to be tested.
-Thermal cameras often used during drought to understand if the plant is receiving too much light.
-Image Aquisition
-I could look at having one robot doing all the spotting and mapping while another does the spraying.
-Used John Deere as a Case study example of Repair and Service fault where farmers are not able to try and fix their own machinery and have to rely on the service from John Deere themselves.
-Key to separate the modules clearly inside the device so that it is easy to repair and understand.
-Plug and play architecture
Suggested tech to use:
-Recommended that I have 2 seperate 'brains' - one for the driving of the device like an Ardupilot. (The same tech that was suggested by Daniel Harmsworth) - the other one for the decision-making around the images the device is recieving. These two brains will be able to communicate with one another.
-for the imaging 'brain' he recommended I look at something like a Raspberry Pie, however this might struggle a bit. Otherwise a WIFI card like (Doccer?) which allows for hardware to hardware transmission, meaning the 'brain' inside the device can communicate with the computer inside the house which carries some of the load in the decision-making process.
-ROS is also a very popular choice and which is currently used often in agricultural contexts. The major difference is the learning curve is quite steep.
The cost of the different 'brains' will also be something to consider in how affordable the final product can be for the customer.
Overall I found this conversation with Brett really engaging and found him to be very open about his ideas and what technology is available in this area, which I greatly appreciate.
He has also indicated that he would like to see how the project progresses and is available for feedback and advice at any time.
Andrew Reymer- Grasslandz - 18 May
-Suggested I look up Trevor James from AgResearch, Paul Addison from Nufarm
-They have a particular issue with Yarrow, which is a mat of weeds.
-There is a big emphasis on 'Regenerative agriculture'
-minimising waste- good for the earth and the pocket
-could it pluck the weeds?- different versions
-"information is power" - nitrogen- could it scan for that?
-How it charges- if it could use solar or wind power that would be perceived as good.
-poa- a weed which comes up after pugging in winter- stops rye grass, dies in November and summer grass comes through.
-Some grasses are hard to eliminate and look similar to normal grass.
-Cape thornapple is a common weed to look at.
-multi-spectral cameras could be useful and might be the way to go forward- what are they scanning for? - could it be more than weeds?
-Rental Idea could work
-Cow Halter by Craig Piggott- example of kiwi innovation
- "When people see cows in a field of clover on their milk cartons around the world, they are seeing a representation of New Zealand as that is not the reality in many places."
-Andrew was very interested in engaging further about my project.
Caprece Trail and Family Interview
- Have a lot of thistles
-Own a horse farm
-"I don't like to spray so I do a lot of grubbing."
-Sand country
-fairly rough grass
-They originally lived in Masterton where they encountered different issues with weeds and had a lot of stinging nettle, which even goats would not eat.
-10 acres of land.
-Grubbing is the method that they use to remove weeds - "I find it healthier and I enjoy the walk."
-Plant grass seed as if there is more grass it outcompetes the weed.
-Timing is an important factor in the way that farmers operate.
-She thinks that using herbicide mainly just thins out your grass
-Thistles are strong at the base.
-Needs Large wheels
-Needs Clearance
-Example of some quad bike bars that can bend down fences
-ponytail is a weed that can kill horses and spray does not work.
-Macrocarpa trees
-"If we had a perfect world we would not use herbicide."
-"New Zealand is small enough that we should look at alternatives."
-"I would like to see New Zealand think about the environment."
-Suggests that Cross-grazing is a good way of managing weeds (cross-grazing is when you utilise more than one type of stock grazing the same land, and they eat different weeds)
Samatha Trumper Video Interview- 27 May
-Weeds can be poisonous to horses
-They usually cut them out.
-Many of the weeds that are bad to horses are different in colour so they are fairly easy to spot.
-approximately 80% of weeds are only bad when eaten in large quantities.
-Penny Royal- they smell and are toxic.
-They dont like; carrotweed, buttercup, clover grass and ragwort
-Sheep eat way more weeds than horses hence why they often allow them to graze similar paddocks.
- "You gotta pull them out at the roots."
-Also suggested cross-graze
-Spot-checking needs to happen often.
-"bright colours are a good indicator of the weeds that are bad."
-If the weeds could be mulched or broken down.
-Could it be powered through electric fencing
- it would have to be able to pull out to the roots
"For horse owners, carrotweed and dock would be the main problem and having something that maintains them would be awesome."
-"You dont want weed in your hay."
Michael Beech- 28 May
-Biosecurity Officer
-Taranaki regional Council
Environmental Officer for 18 months
Previously worked as a town planner
-Pest management up in Auckland before that.- ecological control
-also worked for DOC
-Passionate about pest plants
-Methods of reducing herbicide currently.
-Asked really good questions that allowed me to explain my ideas to him also.
-Showed him my concepts
-Taranaki - lots of flat terrain with steep hills around the edges and towards New Plymouth, which is where a lot of the small holdings are.
- What types of weeds it is trying to target...
-thistles- rosette stage, can grow quite tall.
-Little wand that can move over
Nodding thistle and others can grow quite tall.
-Yellow bristle grass, a big issue in the Taranaki area and moving up to the Waikato. - herbicides don't kill effectively, Dockstar, Proma-S, annual, obvious bristle, 90 seeds per seed head, Cattle wont eat it. - Expensive and long withholding period for the herbicide that makes it a pain to have to deal with. Prevention of spreading is key, dies off with the frost. Does not compete well with other grasses, commonly grows on the roadside and is spread by mowers.
-Reducing cost through the device.
-Mechanical removal such as pulling it out.
-Push prevention
- Flea-bay- starts off as a rosette and grows up fast. Thistle seed-head that blows in the head.
- Giant buttercup
"Your buggy could come in handy for the giant buttercup... that plant can develop herbicide resistance if you keep using the same herbicide. "
-Rhizomes- the idea that there are spores through which some of these plants are able to grow so if you can deal with them then that would be good. -Californian thistle and giant buttercups.
-Velvetleaf- Linked to Sally Linton - seed viable for 60 years in the soil. Current method is hand removal
-If there could be something that could go into crops that would be beneficial.
-Mentioned Trevor James
-"Traditionally that has been the easiest and most effective way of dealing with weeds has been herbicide."
-Selective herbicide, flumetsulam, 2D4D
-Mechanical removal- have you looked at that much?
-flame throwers- fire makes gorse worse because it bursts its seeds open.
-discussion around if
"maybe its a matter of leaving the plant to rot out on site as well"
- if you cut scotch thistle then it struggles to come up again. - buttercups it wont work for.
"It is always a challenge because some methods work for others"
-stubborn and resistant weeds
-docks - easy detection
-no native thistles in NZ- easy to study and easy to target.
-"a drone is a drone" - 20L capacity at the moment
-Different herbicides- granules, herbicides. - Need to factor in what people use.
-Granules- often mixed with water, or can disperse around or into the crown of the plant.
-Ganura- chilean rhubarb. -Ornamental import that started growing wild.
Wild ginger
-Environmental Weeds
-pre-emergent spray is broad-casting
- "the cost of herbicide to farmers is a lot of money and it is obviously also money and time and labour going weeds... I see some really positive things in your project there."
- different climates have different weed types
-drier areas- narsela tussock
whorehound, hawkweeds,
dairy- docks, buttercups, thistles,
-"Some plants will thrive in acidic soils, so you will see more of a prevalence of one type of weed in one area as opposed than the other."
-wetter areas- creeping buttercup
"Climatic factors are also a big thing as well about what weeds will thrive in a particular area."
-Climate change is going to effect where weeds rock up
Ella Patrick effect- 'poisons' the soil around it so that it can have a pure stand and not have any competition. - Wooly Nightshade. Growth inhibitor.
-"With herbicide resistance, if people aren't educated in that sort of thing they thrash say giant buttercup with it and then they say 'ah I keep spraying it and nothing is happening."
The weeds growing faster than the weeds.
- The holes in the soil can leave space for new weeds to come up.
-Suggested looking at weed wipers- if the device is bumping around the weed would that not just spread the seed more?
-stump post-gel. - snip the weed and then targets the weeds. Labour intensive. woody weeds this option could be feasible for.
-Timing- understanding their lifecycles so try and catch them before they flower.
-Seeds and weeds readily attaching to machinery and clothing, this will be different seasons for different weeds.
-weeds can also spread vegetatively.
- "Farmers know when the best timing
-Keen to keep in touch.
Rod Interactions
I have had the pleasure of interacting with Rodney throughout the semester about the state of my project and he has provided incredibly useful information throughout about I could explore and alter to express myself in a more designerly manner.
Lyn Interactions
Daniel Interactions
I have had the pleasure to interact with Daniel Harmsworth extensively throughout this project. Not only does he have a wide skill set and passion for this area, but he also did some work in the agricultural sector before and so has an understanding of what has been explored before. The conversations we have had have been incredibly insightful and have served my project well.
Survey Participants Emails
Thank you for taking the time to fill out my survey on weeds and for listing your interest in further chatting with me about your experiences on the topic of weeds.
Sorry, I noted your confusion on the question about geofencing, this was to determine if people had facilities like these on their farms to map out paddocks, as one of my concepts deals with a device that can roam around paddocks and spot-spray weeds.
Who am I?
My name is Zene Krige, I study Industrial Design at Massey University in Wellington and I am in my final year of study and want to design a device which is able to spot-spray weeds.
My aim is to create a device which is approachable and easy to use by lifestyle block owners as well as farmers and which utilises current and emerging technologies to offer benefits in productivity as well as connecting to some of the systems that may already be existing on farms.
I have done fairly in-depth research into weeds, their causes and herbicides that are recommended to solve them, but I want to gain a better understanding of weeds and what your user experiences with them are. There is a lot of work happening as part of my research including a report that is not accessible to anyone yet, so if you feel that you wish to see some of this work feel free to ask.
If you would like to have a video conference call, then feel free to let me know and we can schedule a time that suits you best, it is always nice to put a face to the person you are speaking with.
Questions: Please answer these questions as thoroughly as possible, I appreciate any and all insights. If you feel you need clarification on any of these please do not hesitate to ask.
1. What types of weeds do you have on your land? If you do not know, or even if you do, please feel free to send some photos of these through to me.
2. How often do you have to treat the weeds?
3. What type of herbicide do you use? Do you think it is working for you?
4. How long do you spend on average per week/month/ year on weeds?
5. Do you treat the weeds yourself or employ a contractor?
6. What are some of your frustrations around weeds and the tools you use to get rid of them?
7. Do you think that branded products factor into your decision of which tool to buy?
8. What brands do you identify with?
9. How important do you feel it is to buy locally?
10. What are some of the key characteristics that make you likely to purchase a piece of machinery?
11. Could adding a fertiliser option be possible?
12. Do you do any other activities in relation to weeding, for example, do you mow the grass first, fertilise soon after…. Etc.
13. For this question, please see Page 1 of the attached PDF and let me know which existing products you like and dislike, and why. If none, could you please explain why or offer me a suggestion of something that you do like?
I have also attached some of the sketch thinking I have been doing around one of the particular product areas.
I understand that we are going into Level 2 which will be a busier time as we all adjust to a new normal, but any feedback, critique and insights would be greatly appreciated.
I look forward to hearing from you.
Kind Regards
Zene Krige
Acknowledgements
I would like to thank everyone who has contributed to my project this semester, in particular during such a difficult and new way of operating due to COVID-19. I would like to thank the lecturers for their continued support despite the distance learning. I would like to thank the technical expertise of the workshop at the very start of my project, and I look forward to seeing them again in Semester 2. I would like to thank my industry experts for their input and valuable technical feedback which has helped shape my understanding of the issues around weeds in New Zealand.