Robotic harvester plucks an Apple every 7 seconds
freshplaza.com
freshplaza.com
For instance, this apple harvester is using UR5 robot arm. It's a very research-friendly robot that I like to work with, but it's nowhere ready for outdoors.
Its reach is just shy of 1 meter -- not practical for apples as well.
Their choice to place an Intel RealSense camera on the wrist of the robot means that they have a USB cable running through the robot. My experience shows that even heavy-duty cables only work for hundreds of hours in such a mode and then require a replacement.
The soft gripper that they use is a good choice for the task, but it depends on compressed air, which means an extra tube running through the robot arm - usually it means even more maintenance required.
Still, this approach can be productionized, but the end result might look very different from this prototype.
Putting the processor at the end of the actuator or sensor arm puts large stresses and vibrations on the electronics, I'd avoid it.
Industrial and armoured USB cabling exist, including 360 degree couplers[1], but obviously it is specialised. Industrial Ethernet is better (e.g. GigE Vision) from an integration pov.
(I remember reading about cultivating trees in a way to make it easier for humans to pick, and something about making it easier for robots to pick. The downside is that it requires a lot more effort at pruning and maintaining the trees, when adjusting for wind, sun, and water).
Only once everyone is using robots for apple picking, we might see some new orchards deciding on a denser layout / higher trees.
Does that apply to cables designed for use in robots? eg https://okielectriccable.com/okis-robot-cables/
The problem with USB is that the protocol is designed with no data loss in mind. A single bit flip due to EM interference or bad cable, and the camera firmware (or the host driver) hangs and require a restart. Intel RealSense and Microsoft Kinect Azure cameras all have to be plugged into USB hubs which allow power cycling. Otherwise, it's just not practical. And even then, after a few hundreds of hours the frequency of restarts it degrades even further.
I might sound a bit dramatic, it's mostly that my scars are too fresh.
USB employs two error checking methods to ensure that data is sent correctly:
https://web.archive.org/web/20130615145009/http://www.totalp...
In Ethernet, on the other hand, packet loss is routinely handled, and higher level protocols can work when physical media is less than perfect.
> If there is a data error and the CRC check fails, however, the receiver will not reply with an ACK, and the transmitter is required to resend the data with the same toggle bit. The transmitter will continue to resend the same data with the same toggle bit until the receiver ACKs its reception.
Then my explanation of the observed phenomena is likely wrong either partially or fully.
Farm robots always seem to point out to me the 'white tower syndrome' that researchers fall into. Dirt and robots is like the old joke about spherical cows.
Guess what, plants grow where it rains consistently. Plants therefore grow in mud, and sometimes we throw poop at them on purpose. Sooner or later it will start raining just as you are at the farthest point on the property from shelter, and early season work often coincides with freezing rain or hail.
And the more you avoid working in the wet, the more you have to deal with dust and grit.
Incidentally, part of my evolution toward gardening in the rain is realizing I'd rather be wet than sneezing, coughing, or getting dust in my eyes. Rain is also more pleasant than sweating like a pig. Luckily I'm not dealing with clay so I can bend some rules about digging in wet soil.
Yes. Nor is it the first. Here's one from 2019, and much faster. [1] Another academic one from 2017.[2] A second one from 2017.[3] Another one from 2016.[4]
Eventually, someone is going to get this right. Someone who's good at machinery design. Most of these have dangling wires or duct tape showing, or are far too slow. A good practical question is "Can you pressure-wash it?" This also applies to food-preparation robots.
The computer vision part seems to be a solved problem. All of these can find apples visually.
Few people are really good at mechanical design. I've mentioned that I restore old Teletype machines. Those are very well designed. All the parts are replaceable. They're reliable, even though complicated. They have few wear problems. Very few of the parts are fragile. They are unharmed by minor jamming problems. They do need annual lubrication. There's a complex balance of complexity, manufacturability, maintainability, and cost in mechanical design. Two people, Walter Krum and Ed Kleinschmidt, designed all the good ones of the 20th century. Many others tried, and produced inferior machines.
This is more recognized in the gun industry. There are few good new gun designs. Every few decades, someone makes a good new design, and that gets copied and modified. Well known examples include the 1911A1, by John Browning, the AK-47, by Mikhail Kalashnikov, and the Glock, by Gaston Glock. All demonstrate a mastery of the materials used and a good understanding of the production processes. It's not just the form.
[1] https://youtu.be/aijzVv6UeLQ
[2] https://youtu.be/SwE-LPS01lk
If memory serves, they achieved speed by reducing harvest %.
Sign up for our daily Newsletter and stay up to date with all the latest news!
and two options: "Subscribe" and "I am already a subscriber"
That's like the "No, thanks, I am stupid" dark pattern, but somehow more stupid and less dark.
And the thing is that orchardists already have expensive harvesting equipment. Tech over and over again likes to fix 'greenfield' problems and then act like most of their target audience isn't 'brownfield'. If you wait long enough that becomes true, but you get a lot of nasty surprises that should not be surprising at all in the near term.
A caterpillar picker or an aphid squisher might be a better use for automation. Hell a caterpillar spotter that can classify species without even touching them might be useful enough.
For juice/cider/etc though...
Some people think the latter is part of the obesity epidemic. We have lots of cravings when we are malnourished, and while hyper palatable food can explain a lot of this, it's not the only culprit.
Those peaches you eat are the only ones you can get because the others bruise too easily. If you want a 'better' peach you're going to have to plant a tree. It's one of the 'sales pitches' that gardeners and permaculturists use to try to convert people. If you want more nutritious vegatables, you're going to have to get your hands dirty. Vine-ripened foods can be a profoundly different experience than store bought, as anyone who has ever grown a tomato will be all too quick to tell you.
When I first moved there I was absolutely shocked by how good the fruits/vegetables and even corn tasted.
Shouldn't be that hard to improve on though, it looks rather crude compared to cutting edge robot tech in other verticals.
https://research.qut.edu.au/future-farming/projects/harvey-t...
"In early 2017, worldwide literature indicated a success rate of only 6% in testing scenarios similar to those used for Harvey, and up to 30% when the crop is modified and leaves are removed."
Picking one apple every 7 seconds is not bad, I'd say.
I've seen restaurant owners hire folks to hit on them in particular. Maybe an industry thing? Music / film ideally doesn't have that issue but less exposure to that world.
I'm interested in the compressed air technology though. I remember Disney experimenting with that as part of an effort to make their animitronics safe for interaction with people. There are also some things I know from internal martial arts that I wonder about applying to robotics.
They don't even let you type the words App Store without auto capitalizing it for you even though they are not the only app store and they really don't have a trademark on it.
Articulated arm robots are neat but it take a lot to make them efficient.
But despite this cost savings, the prices for food keeps on going up. Who’s absorbing all that extra value?
Robots, unfortunately, move capital from labor to owners.
I'm not proposing robots are bad. You're kinda screwed once they're out in the wild, if you don't adopt them you can't stay in business eventually (without some sort of pivot to a niche market segment).
Areas where there is automation but prices continue to go up have issues with competition - banks are a great example because banking licenses are impossible to acquire.
I would even add in for automation for cars or heavy manufacturing: they increase the cost of getting into the game: thus driving down competition as well.
The automation started in the late 70s at least tho. The other child comment about banks and lack of competition I think really hits this on the head.
How can automation add value instead of capturing existing value through better efficiency?
I once heard this example using coffee beans. If you buy coffee beans, and you have to brew it, that's a commodity. When you go to a coffee shop, and the server brews it for you and brings it to you, that's a service. When you have one of those single-server instant coffee that you pop into the machine, and it brews it for you, that's a product.
The distance between you and the original commodity can be defined as the number of transformations from the coffee plant to your cup. Raw beans that you have to process yourself is closer to the original plant. A service provider inserts themselves into that process of brewing a coffee, and charge you for it. Productizing it so you don't even have to wait on someone else to do it for you inserts additional steps -- packaging, technology, etc.
We call this "value-added", meaning that some merchant captures the value from transformational steps that you would have otherwise have to do yourself.
However, I am talking about "value-adding". The key difference here is that the transformation changes the commodity in a way that contributes something beyond what you could have done yourself, not simply capturing existing value in the value chain.
I think most use of technology and automation are attempting to capture existing value rather than creating value that was not there before.
Network effect of the internet might be value-adding, but we have also seen the same network-effect used to aggregate and capture large swaths of the market without contributing anything back (Facebook, for example). Is the network effect of Amazon's logistics prowess value-adding? And if it is, what unintended consequences have we seen from it?
It may also be that, for computing, we would have to look at older generations of our software and look for the forks not taken. As an example, Apple's Hypercard enabled individuals to create their own computing experience, something that could not have been done before without it and other systems like it ... and it was deliberately dropped so that all design came from Apple (iPhones, iPads, etc.). By the metric of return on shareholder value, that decision was wildly successful. But in terms of adding value for our civilization, was it?
AlphaGo and AlphaGo Zero demonstrated that humans can no longer beat AI Go players. If you are willing to spend energy to power all those TPUs, they can play far beyond the best human players.
That shook up a lot of the pros. Some of the top players have quit. However, other players got more excited. They saw Go being played in ways they had never seen before. People who made their living tutoring Go to other people have lost their livelihood. On the other hand, Isee discussion at the high amateur ranks now uses Leela Zero, or Katago to learn and explore the game. That's a positive contribution that could not be done before. (And the shadowside: it is easier for people to cheat; though if you are using Go to develop your own intellectual ability to evaluate and make decisions, using an AI to play competitively is kinda cheating yourself).
The difference between today's computing culture and the "road not taken" you described is like the difference between a scribal culture and a literate culture. I really hope the ship hasn't sailed.
I am not sure if that is a good analogy. Maybe it is.
By the “road not taken” though, I mean something that is more difficult to pull off, and it does not monetize well, if ever. I am thinking, for example, Christopher Alexander’s work on pattern languages. That is, designing a pattern language in such a way that the end users can always create the right thing for themselves. Something like the ideas behind Smalltalk, where you don’t need a computer science degree to customize. Instead, it got turned into OOP Gang of Four. A smaller artisan class of software engineers and product managers to determine what the best design for the end user. Add A/B testing, growth hacks, dark patterns, and customer validation, we have the engine of growth for startups to return multiples for investors.
I grew up in the 80s and 90s where the magical experience of computing had a lot yo do with creating those environments.
Now, flip side. I have also read a history of MMOs that, the author concludes that the all time most successful MMO is Roblox. Players make games for each other, and there are some revenue sharing. There is a pattern language in the design, but otherwise, there is a great deal of freedom to make stuff. It still netted a good return for shareholders.
So maybe that ship has not sailed yet.
But to come back to that question I posed: is Roblox value-adding? If so, what is it contributing that could not have been done without it?