Robotic Following Cart to Replace Light Duty Tractor Tasks?

   / Robotic Following Cart to Replace Light Duty Tractor Tasks? #151  
300 pounds is fine if it fits the product. Like lettuce...but I think we are worried about you trying to haul apples with this thing and
you hit a nice root with a full load on a three wheeler and you might have some excitement. Two grippy tires pushing/pulling I think are fine.
but for a stable platform, four legs are better than three. Six are better than four...and by then we have a eight million dollar moon rover before
all everyone's suggesting gets incorporated...;)

This actually may be a fifteen thousand dollar product that you are trying to bring out at too low a pricepoint. It's all about ROI for a business.
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks? #152  
Having returned from my older brother's home in Virginia in the wine country, where every turn in the road a new super deluxe winery is hidden up on rolling hills, I can see
a market for this "aide" being bought by very wealthy landowners who dabble in wine/have made it a business. So paying 15K for something that helps them and makes them look good/advanced, always an important issue to some, may be pocket change. Look what tractors and implements cost. I'm not saying price your cart like a tractor, but the folks who buy these things for business applications buy other equipment that costs much more. Again, if one can show ROI, while I know you have already had price creep, maybe it needs to creep a little more.
A stable platform is one of the first things I'd look for, and then safe operation.
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks?
  • Thread Starter
#153  

Short video from testing today. We are still primarily testing software - what mechanical/abuse tests would people suggest?

Assume:

(1) 300 pound limit up/down steep grades
(2) mud/ditches to see performance
(3) range testing in cold weather

Others?
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks? #154  
Put a shovel handle on the ground 45 degrees from perpendicular to the path of travel and see how it goes over that with and without load.

Several angles, for that matter.
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks? #155  
You may have already mentioned this but does the operator have some kind of transmitter device?
What is to keep this thing from following your 100 pound dog out of the orchard?
What happens when it is following someone and someone else walks up near it, in front of it between the cart and operator, perhaps all
lumped into radar issues. Have it be following you and call a dog to come between the worker and the cart. (after you are sure of course the dog won't get hurt...)
Shovels, rakes, all those things called in boating hazards to navigation.

And try running over some plastic bags, whatever you can think of someone shouldn't have left on the ground in your target markets.

Greenhouses are wonderful for flat floors. Real orchards as you know well are anything but flat. I'd drive it into varying size holes also.
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks? #156  
View attachment 529975View attachment 529976View attachment 529977

Those pictures show where we are trying to go with it for version 1.0. Do you think we need 4WD in those types of settings from the images? We are looking at 300 pounds of capacity, 2WD drive with a rear caster, 15 miles of range, autonomous without any beacons/infrastructure etc., and around an $8K price through a dealer. On paper this would be about a 2 year payback in many berry operations (and much shorter if labor goes towards 15-20 bucks an hour which we hear guys are actually paying today to keep people harvesting).

In a crop like blueberries, a single picker can pick 20 pounds / hour. 300 pounds of capacity is thus 15 man hours of picking. If we go much larger the product becomes much more expensive and needs to run enough to support perhaps 20-30 guys picking. This is why we have settled towards the 300 pound weight capacity for version 1.0. Does that make the rationale sound more logical?

Thanks for the feedback.

Your best bet for being able to sell these things, is to make it as widely diversified as possible. Instead of thinking about having it only run down perfectly straight rows, with perfectly level ground, and no debris, roots, rocks, holes, sand patches, legos bricks, whatever, you need to make this thing adaptable to irregular terrain and conditions. Else wise you will be excluding a sizable percentage of your potential "market" for this product. Your potential buyer needs to feel confident that not only with this thing navigate THEIR existing terrain conditions, but that it will be able to do so while fully laden (and even a bit over loaded, because these things happen).

I have serious doubts that taking that one center caster, and replacing it with a caster at each back corner, is going to skid your entire production future into the ditch.

But what it will do, is allow a potential customer to see that maybe, just maybe this machine will survive operating on their specific site, in their specific environmental conditions. As another poster (or more, long thread) has said, 4 legs, or in this case 4 wheels, is always, always more stable than 3. Especially with something that is going to be carrying weight around. And sometimes that weight will not be evenly distributed, or it will be stacked high enough to raise the center of gravity. Experience teaches us these stability "lessons" in life, and sometimes it does so the hard way.

I see a whole patch of potential customers taking one look at this thing, with it's 3 wheel (one being a much smaller caster swivel) arrangement, and just saying "Nope", and moving on. I have to ask, why are you so determined that you MUST have this one centered caster on this machine?

4 wheels, one at each corner, all 4 the same size. Make the 2 on one end swivel casters if you want, and the other end has the drive wheels. Fine. I'm not saying that each wheel MUST be motor driven (4 wheel drive), but you NEED a full size wheel at each corner. I do think a true 4 wheel drive unit will be more widespread in popularity. Also a full track model might actually be better and cheaper (than 4 wheel drive), as you don't have to sort out steering for 4 wheel drive, and can still drive it with only 2 drive wheels.
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks?
  • Thread Starter
#157  
You may have already mentioned this but does the operator have some kind of transmitter device?
What is to keep this thing from following your 100 pound dog out of the orchard?
What happens when it is following someone and someone else walks up near it, in front of it between the cart and operator, perhaps all
lumped into radar issues. Have it be following you and call a dog to come between the worker and the cart. (after you are sure of course the dog won't get hurt...)
Shovels, rakes, all those things called in boating hazards to navigation.

We have some really cool machine-learning-based code for this and a vision based approach. There is no transmitter.

We have not tested near many dogs, although we have tested in a lot of very windy foliage and around many people (it locks on to one person).

We'll work through your scenarios.
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks? #158  
This is why I love this forum! The cool **** y’all come up with amazes me!!!!
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks? #159  
I built a 3 wheeled box to haul wood in. It only goes on a hardwood floor and is still not as stable as I had in mind. It works for the intended use but I have to load the 2 wheel side first. I think your cart would be a lot better with 4 wheel. Now, if it had a claw and could stack firewood it would be great.
 
   / Robotic Following Cart to Replace Light Duty Tractor Tasks?
  • Thread Starter
#160  

That quick video shows the machine in use a bit more. Not loaded up yet - we'll get to that soon. Trying to test autonomous behaviors first unloaded as everything changes with load and it is easier to code around this if behavior is perfect first unloaded.

We are going 2WD at the moment because we don't think a typical customer will want to spend another $1K on top of $8K or so for another set of motors/tires/axles. We will certainly add this later.

It is our belief that in the next 10 years, claws and capability will be there to do things like stack firewood, prune fruit, etc.. We believe our platform is the ideal launching off point, however, because it is achievable today, and can later be fitted with those types of capabilities.
 
Last edited:

Tractor & Equipment Auctions

2006 Ford F-550 4x4 Flatbed Pickup Truck (A50323)
2006 Ford F-550...
2013 Jeep Patriot SUV (A50324)
2013 Jeep Patriot...
2012 Honda Accord Sedan (A50324)
2012 Honda Accord...
Toro Sprayer Cart (A50324)
Toro Sprayer Cart...
2001 DITCH WITCH JT1720 DIRECTIONAL DRILL (A51242)
2001 DITCH WITCH...
2018 Toro Grounds Master 7200 72in Zero Turn Commercial Mower (A50322)
2018 Toro Grounds...
 
Top