I do think permaculture is the most advanced method of farming in that (when done right) it produces increasing yields year-over-year. If high-yield monoculture farming is an industrial approach to farming, permaculture is an information-age approach. But nobody's tried to automate the design work, and for good reason. It's not that that would be impossible - I think a sufficiently advanced sensor drone coupled with a sufficiently advanced machine learning algorithm could design a permaculture installation as well as a human. But then the question becomes, are the supply chain costs of creating the drone and the algorithm cheaper than educating the human and having them do the work, especially given the health benefits of light exercise like gardening? Nobody with the money to do it has tried that yet.

Some friends and I have been thinking that an ideal setting to advance this technology would be on a semitraditional farm, where the researchers spent a few hours a day working on the farm and a few hours thinking about how to automate their most time-consuming tasks. If we do manage that we'll be sure to put our results out there.