Small Robot Company - An opportunity for farmers to own shares in us

[SmallRobotCo]

What about a tractor unit like a sprayer and instead of booms the drone or robots use the (sprayer) as a docking station and as it goes up and down the drones spread out and do the job, so basically you still have tram lines and a load of gremlins which charge up when parked on or hard wired to main unit but are free to cover their own patch, saves the battery issue, just an idea

That's an interesting thought! Worth exploring as an idea...

 

[willy]

That's an interesting thought! Worth exploring as an idea...

Remember where you heard it first

 

[SmallRobotCo]

I admire your goals but disagree with your approach. The digital field philosophy is a grand long term aim but the amount of R&D investment to reach that point will be astronomic and the technical and agronomic advances will take time to develop. In some areas we don't yet know what it is we don't know. Grand philosophies will never sustain a viable business but a great business can work to achieve the aims of such philosophies. The weeder is one of the many tools you need to develop to reach the digital field goal but crack the weeder and use it to generate revenue to fund the next steps.

What is the first piece of kit you see farmers being able to do away with when they lease their first robot? If the first robot I lease is a wheat crop weeder I still need my sprayer to apply fungicides and insecticides or am I missing something? If I lease a weeder to replace a hand weeding operation I instantly reduce my labour and chemical costs. I can tell you it is ten times easier to sell a product to a farmer growing a crop worth £5000/ac than one growing a crop worth £500/ac..

I also disagree about your working window, you will know you can not simply start at one end of the field in October and conduct one pass of weed control across the field by April. There will always be a spring flush of weeds in both a high value and low value crop, you may have a critical 3 weeks in lettuce and 6 weeks in cereals but the higher value crop would more than justify having twice as many robots per Ha. Further to that, fields producing lettuce in particular are generally multi-cropped, the robots can carry out weeding operations over a much longer March to October window. Also consider the lettuce transplant is a considerably bigger plant than the weeds around it, that makes targeting the weeds far quicker than in a cereal crop.

Thanks for the message - I think where we disagree is that I see the digital field as being the starting point. Get that in place, and the highly accurate weeding and in time highly accurate everything else becomes possible.

And I don't think it will be that expensive - in fact we can already do it very cheaply. We need to increase the speed with which each robot covers the ground definitely (it's still early days on that front) but I'll share a video shortly showing the quality of the images we are getting. We can already identify an individual plant in a field, locate that plant to within 2cm, and tell whether that plant is wheat or not.

We then need to do that at a larger scale.

Understand the reservations about wheat as a first market. I've laid out my thinking around that already so I won't repeat myself here. NB I didn't say you would complete the weeding in a single pass. Why couldn't it be lots of passes - a constant weeding job with lots of weeding robots?

 

[Grass And Grain]

On the point around "why not on a tractor", because tasks such as non-chemical weeding are much better suited to an autonomous smart machine. If it reaches a patch of heavy weeds it will slow down and patiently deal with each weed. A bloke on a tractor wants to drive at the same speed up and down the field, which makes it more difficult to get the technology to work properly.

Couldn't you just use a diesel electric tractor, driven autonomously, then it can go as slow as you want.

I do get whay you are saying about heavy tractors wrecking the soil on headlandd though.

Imteresting concept. Good luck. It needs A LOT of thinking about as to how the whole system would work before spending too much on development imho.

 

[SmallRobotCo]

But why would you start with a tractor? Is the tractor the best tool to do the job as accurately as possible? If not then we should design something different, rather than being fixated on what we already have.

My feeling is that tractors have been designed to be as quick as possible (to maximise the efficiency of manhours, as others haves said in this thread), not as accurate as possible. We need to change our mindset.

 

[gavztheouch]

I am also developing a small land drone to selectively spray weeds. Its just a hobby project but im hoping it will attract others to work together to hopefully design something that farmers can build and maintain themselves. There are a lot of companies trying to do this right now it will be interesting to see what approach works best

https://thefarmingforum.co.uk/index.php?threads/autonomous-rover-project.249634/

Recent advances in machine learning have made this possible, detecting a weed in a image of crops is no longer a hard problem. Exciting times

 

[SmallRobotCo]

I am also developing a small land drone to selectively spray weeds. Its just a hobby project but im hoping it will attract others to work together to hopefully design something that farmers can build and maintain themselves. There are a lot of companies trying to do this right now it will be interesting to see what approach works best

https://thefarmingforum.co.uk/index.php?threads/autonomous-rover-project.249634/

Recent advances in machine learning have made this possible, detecting a weed in a image of crops is no longer a hard problem. Exciting times

Thanks - looks like a really interesting project and it would be good to have a chat to see if there are some ways of working together. Farmer led, farmer designed with really clear positive outcomes for farmers is key to all of this being successful

 

[SmallRobotCo]

An update on our Crowdcube campaign - we're launching tomorrow!

We've had such a successful week on in private mode on CrowdCube that we are going to bring the page live tomorrow.

You can get early access to the site here crowdcube.com/smallrobot

 

[SmallRobotCo]

As mentioned in this thread, our Tom robot has been working with the Waitrose Farm down in Hampshire. Here he is in the field - and I'll send through

 

[SmallRobotCo]

Use subtitles on this one - it was a bit windy!

 

[SmallRobotCo]

 

[SmallRobotCo]

This video shows our data gathering robot in action on the Waitrose Farm. It's in a grass / clover mix in this example, but the important thing is to look at how detailed the image is that we are getting as a result.

We're not expecting farmers to look at each one of those thousands of images - the AI algorithm will analyse the images and then show the farmer on a map exactly where each weed is or each plant is. This is valuable in itself, but this will then be the foundation data that we use to make the other robots work effectively.

 

[curlietailz]

Made a small investment.... I think your idea is great
And do see swarm-bots as the future
Good Luck

An update on our Crowdcube campaign - we're launching tomorrow!

We've had such a successful week on in private mode on CrowdCube that we are going to bring the page live tomorrow.

You can get early access to the site here crowdcube.com/smallrobot

 

[curlietailz]

I see it’s usefulness for punch planting into stubble.... although it would need someone to fill the little hopper at either end of the field? And for post emergence weeding....in cereal and grassland weeding
But it wouldn’t work once the crop starts growing a stalk in the spring

 

[SmallRobotCo]

Made a small investment.... I think your idea is great
And do see swarm-bots as the future
Good Luck

Thank you so much for the investment! I really appreciate your support for what we are trying to do.

 

[SmallRobotCo]

I see it’s usefulness for punch planting into stubble.... although it would need someone to fill the little hopper at either end of the field? And for post emergence weeding....in cereal and grassland weeding
But it wouldn’t work once the crop starts growing a stalk in the spring

Not in it's current form it won't, you're right - we need to do a re-design of the frame over this winter so that it will travel over the wheat throughout its growth cycle. That's one of the things we are raising funds for.

 

[Grass And Grain]

Do you think you will be able to get the robots accurate enough to take out weeds that are growing intertwined with the crop plants?

 

[SmallRobotCo]

Honest is I’m not sure on that one. My immediate thought is yes because we will be able to identify the individual plant and then touch the plant we want to kill with an electrical charge. But it’s a good question - not one we’ve been asked before - will check with the tech team!

 

[Devil's advocate]

How many lettuce farmers are there in the UK?

Not going to sell many units.

Veg industry employs more staff many times over compared to broad acre cereals/livestock.

salad & field veg massive industry such a range of crops but mostly in rows. Same tech then leads to sugar beet, maize, oil seed crops & potatoes. Crack that cereals/grassland then easy.

 

[carpenter1]

Thank you so much for the investment! I really appreciate your support for what we are trying to do.

I like what your doing, when do you visage making a profit?