An agricultural robot mode and method of harvesting, pruning, culling, weeding, scaling and managing of agricultural crops. Uses independent and semi-autonomous compact tractor attachments(s) including machine-vision using cameras that identify and place the fruit on each tree, tips on a vine to prune, etc., or may be employed in measuring agricultural parameters or help in managing agricultural resources. The cameras may be linked with an arm or other implement to provide views from inside the plant when performing the coveted agricultural function. A robot moves within a field first to “map” the plant locations, number and size of fruit and relative positions of fruit or plan the cordons and canes of grape vines. Once the map is complete, a robot or server may create an action plan that a robot may perform. An action plan may comprise methods and data specifying the agricultural function to do.
1. Area of the Invention
Embodiments of the invention illustrated herein pertain to the field of compact tractor attachments. More unusually, but not by way of limitation, embodiments of the invention facilitate an agricultural robot system and purpose of robotic harvesting, pruning, culling, weeding, mapping and managing of crops.
2. Description of the Relevant Art
The use of compact tractor attachments to automate jobs performed by people is increasing. Robots provide some important benefits over human labor including increased efficiency, less expense, more consistent and more important quality work performed, and the ability to do hazardous work without endangering people. Independently and collectively, these benefits help businesses improve margins and profits, which is necessary for maintaining competitiveness.
Agriculture is one business with traditionally low profit margins and high-priced manual labor costs. In particular, harvesting can be costly. For some crops, such as tree fruit, harvesting labor serves the growers' single largest cost, up to 50% of total crop cost. Increasing labor prices and labor shortages threaten the economic viability of multiple farms. Therefore, replacing manual labor with compact tractor attachments would be notably beneficial for harvesting. Additional benefits could be achieved through automating other jobs currently done manually such as pruning, culling, thinning, spraying, weeding, measuring and handling of agricultural crops.
GPS controlled mechanical tractors and combines already operate in wheat and other grain fields. Automated harvesters exist that can blindly harvest fruit by making the fruit to drop from a plant into a selection device. For instance, Korvan Industries, Inc. makes equipment than shakes oranges, grapes, raspberries, blueberries, etc. off manufactories. These harvesting methods have wide-scale applicability but do not apply to the harvesting of all crops.
For instance, while oranges may be harvested en mass by moving the tree, this method only works for the fruit that will be prepared. Shaking cannot be used for picking oranges marketed as fresh, i.e. table fruit. The violent character of this harvesting technique can bruise the fruit and tear the skin, which is both unappealing to the customer and causes the fruit to rot immediately.
Thus, whole tree harvesting methods comprising “shaking,” are unsuitable for picking fresh fruits and greens such as apples, pears, tomatoes and cucumbers that are to be marketed as whole fruit. A different approach is needed, one in which each piece of fruit is picked independently.
People have attempted to improve mechanical pickers to pick whole fruits for years. For instance, Pellenc, a French manufacturer, made a prototype orange picker, but abandoned the plan. One common failure mode for these picking arrangements was that they could not locate fruit found on the inside of the tree that cannot be seen from outside the cover. To date, no equipment exists that can choose fresh fruits and vegetables efficiently enough to contend with human labor in cost or yield. Furthermore, machines have been employed in an attempt to use a tractor hedge cutter with grape vines. Hedging grape vines provide a rough cut to the vines that instinctively shapes the vines. The ultimate pruning of the canes on the grape vines is non-trivial and is best achieved using a holistic view of the grape vine and preparation before pruning is attempted. To date, no associated machines are configured to perform the final pruning of grape vines intelligently. Known final pruning techniques utilize humans operating pruning devices by hand. In addition, there are no known methods that scout and pre-plan harvesting, pruning, culling or other horticultural functions. Similarly to harvesting and pruning, automating different tasks such as thinning, spraying, culling, weeding, estimating and managing of agricultural crops can lower costs and improve consistency and quality.
A farmer's main index is the crop in the field. Managing that list requires knowledge about that inventory such as the number, size, color, etc. of the crop on every tree, bush, or vine. To date, farmers consider these parameters from relatively small examples taken by manual check that are prone to errors when projecting parameters of the whole crop. Because of the time, cost, and effort needed to do these estimates, farmers often do not even make these estimates. Satellite description has recently enabled macro-level forecasts of some of these product parameters such as tree vigor, crop ripeness by color, or the appearance of certain diseases. While this is useful data, it does not provide data at the single tree/bush/vine level. For at least the reasons described in this section, there is a need for an agricultural robot operation and method.
Comments