Augmentation is on the road to automation
“We wanted flying cars, instead we got 140 characters.”
The perception about the impact of AgTech over the last few years can be very well summarized by Peter Thiel’s famous line
A few weeks ago, I visited some sweet potato farmers in Merced County, California, who are looking for some harvest automation solutions. They were using a mechanical harvester which has been modified many times over for harvesting sweet potatoes. The harvester still needs people to manually sort the potatoes based on the quality and size of the sweet potatoes.
The current solution requires about 6 people working on 2 rows at a time, and a tractor driver to pull the mechanical harvester. The process is painstaking and a crew of 7 can harvest about 0.6 acres in an 8 hour shift.
They have been using their current mechanical harvesters, which were built in the 1980s, when “Don’t stop believin” by Journey was a top song. These growers have “not stopped believin” that they can get a profitable business, if only they can take care of some of their cost issues.
It is not new news that labor costs have skyrocketed in California over the last few years. These growers are struggling to improve the efficiency and productivity of their workforce to get more out of less. (which has been the ethos of farming and will continue to be so in the future).
They have modified the original equipment so that the sorting and packing crew does not have to constantly bend to put sorted potatoes in a large bin. It reduces bending related stress for workers and makes them more productivit. They have modified the rails on which bins move so it is easier to unload the full bins (which can weigh up to a 1000 lbs) and improves worker safety, reduces tiredness, and improves productivity.
A modified mechanical sweet potato harvester from Merced County California with empty stacked bins in the back
These examples are not about automation but augmentation to assist the human being with small changes.
In 2022, The Mixing Bowl did a study of the agriculture robotics landscape.
For the purposes of this robotic landscape analysis, we focused on machines that use hardware and software to perceive surroundings, analyze data and take real-time action on information related to an agricultural crop-related function without human intervention.
As can be seen from the chart above, vision-aided robotic pickers and vision-aided spot sprayers are some of the hardest problems to solve within agriculture robotics.
Current adoption of automation in harvest and harvest related activities is low within specialty crops.
Primarily, the study finds that the overall advancement of harvest automation in the fresh produce industry is so far limited, mainly due to the technical difficulties in replicating the human hand to harvest delicate crops.
Moravec’s Paradox
Moravec wrote in 1988, "it is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility”
Due to Morevac’s Paradox, harvest technologies are not as far advanced as pre-harvesting, and harvest assist activities as can be seen from the chart below
The question then becomes: “Should robotics, AI, and ML focus exclusively on automation and replacement of human tasks as performed today?”
It is tempting to think automation is the holy grail to solve problems associated with human labor, and efficiency.
But as economist and technologist Erik Brynjolfsson says,
“A common fallacy is to assume that all or most productivity-enhancing innovations belong in the first category: automation. However, the second category, augmentation, has been far more important throughout most of the past two centuries. One metric of this is the economic value of an hour of human labor. Its market price as measured by median wages has grown more than ten-fold since 1820. An entrepreneur is willing to pay much more for a worker whose capabilities are amplified by a bulldozer than one who can only work with a shovel, let alone with bare hands.”
We have seen the development and adoption of collaborative robots (or cobots) as a way to augment human capabilities rather than to replace them completely.
Harvest Assist
Augmenting humans with technology opens an endless frontier of new abilities and opportunities. The set of tasks that humans and machines can do together is undoubtedly much larger than those humans can do alone.
Examples include Burro’s smart farming system. The system is described as augmenting human capabilities and working side by side with farm workers.
“We've built a smarter farming system using user-friendly, autonomous robots that work side-by-side with farm workers to make agriculture more productive and sustainable.”
Last week, I visited some strawberry growers in California during harvest season. Harvesting is an extremely manual process, which requires workers to walk through strawberry rows, determine if a strawberry is ripe or not, pick the strawberry, and put it in a clamshell.
If no machines are used, pickers walk within rows, bend down to pick strawberries, and when their case is full, they walk back to the end of the row to deposit their case/cases, get the strawberries checked by a checker (not in all cases), and record information about strawberries picked. (For payment purposes, as they get paid by a piece rate). The walk back to the end of the row is not productive time for a picker, and it increases the amount of distance they are walking in non-productive activities and increases fatigue for the pickers.
Moving platform ahead of the picking crew to reduce fatigue and walking distance / time for workers (Image by Rhishi Pethe)
Some harvest assist solutions consist of a moving platform ahead of the pickers. When the picker’s clamshell or crate is full, they walk just a few steps to the moving platform and deposit what they have picked onto the platform.
The platform keeps moving to keep pace with the strawberry picking crew. The moving platform, which stays just a few steps ahead of the picking crew, drastically reduces the amount of walking to be done by the pickers. It increases the amount of time a picker spends in picking strawberries (in a given shift). It increases the productivity of each worker, and reduces their fatigue as they are walking less for the same amount of product picked.
Last Friday, I visited another strawberry grower. They raise the beds for growing strawberries, and so a picker has to bend less white picking strawberries.
The height of the strawberry bed in the picture above (Image by Rhishi Pethe) is much higher than a regular strawberry bed.
It reduces the amount of bending required to pick strawberries. The picker’s backs thank this configuration after a long day. The pickers are less tired, and so are more productive, and so they make more money.
What does this all mean?
As a startup or a technology provider, as you think about providing solutions, it is important not to fall in the trap of thinking that a viable solution is only something which completely automates a given process.
There are many humans with high skills involved in farming and it is not possible to replace all these skilled workers. (Picking strawberries at a given speed is not easy!) Most of the growers I talked with don’t have the goal of replacing workers, but the goal is to make them more productive, less tired, and ultimately satisfied and happy so that they continue to work with the grower.
From a selling standpoint, you will have to not only sell to the grower, but also the farm manager, and the workers who work there to adopt your technology.
As you saw with the examples of the sweet potato harvester, the moving platform, and the raised beds, growers are open to any and all suggestions which can help them improve the productivity of labor on their farm, and help them stay competitive in the market.
The only way to find out is to spend time talking with different growers, crew managers, and actual people who work in their fields and not to limit our imagination based on existing configurations and constraints.
Who is Rhishi Pethe ?
Rhishi Pethe is the Managing Partner for AgriFoodTech advisory and analysis firm Metal Dog Labs. He has more than 20+ years of technology experience working within the agriculture and food industries. Rhishi has had product and technology leadership roles at Alphabet X (the moonshot factory), The Climate Corporation (Bayer Crop Science), Amazon, startups (with exits) like HarvestMark and others.
He has extensive experience in supply chain & logistics, product management, data and technology strategy, and artificial intelligence.
He is the creator of the weekly newsletter at the intersection of technology and food/agriculture called “Software is Feeding the World.” He lives in the San Francisco Bay Area.
(image provided by Rhishi Pethe)