I realize that planting isn't the same thing as harvesting, but I once had an acquaintance who was a salesperson for an Australian startup that made automated planting machines. Imagine a robot about the size of a small refrigerator that creeps along the field on bulldozer treads, digging holes and dispensing seeds as it goes.

To try and sell these things, my acquaintance would go out in the field and work alongside the farmers for weeks at a time to try and demonstrate the robot's value. Invariably something would break, and the engineers back at the main office would quickly turn around a design fix.

Although the farmers were usually patient and supportive, rarely did they end up buying. The company went out of business several years ago. I don't mention this anecdote to discourage you from your idea; just to suggest that the problem may end up being a lot more difficult than it may seem.