US 11,665,992 B2
Predictive agricultural management system and method
Sarah Anne Placella, Orinda, CA (US); Adam Ralph Zeilinger, Berkeley, CA (US); Tyler Evan Schartel, Springfield, IL (US); and Ken Yamaguchi, Emeryville, CA (US)
Assigned to Root Applied Sciences Inc., Berkeley, CA (US)
Filed by Root Applied Sciences Inc., Berkeley, CA (US)
Filed on Jul. 28, 2020, as Appl. No. 16/941,217.
Claims priority of provisional application 62/880,224, filed on Jul. 30, 2019.
Prior Publication US 2021/0029866 A1, Feb. 4, 2021
Int. Cl. A01B 79/00 (2006.01); G06N 20/00 (2019.01); G06Q 50/02 (2012.01); A01B 79/02 (2006.01)
CPC A01B 79/005 (2013.01) [A01B 79/02 (2013.01); G06N 20/00 (2019.01); G06Q 50/02 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A computer-implemented method for training a predictive model for crop management comprising:
collecting a first data set about at least one global agricultural factor from a remote data collection service;
collecting a second data set about at least one local agricultural factor from a local data collection device;
assimilating the first and second data sets into a local database to create a first training data set comprising the collected first set of data and the collected second set of data;
training a predictive model using the first training data set to produce one or more measurable outcomes for the crop;
collecting a third data set about at least one global vineyard factor from the remote data collection service;
collecting a fourth data set about at least one local vineyard factor from the local data collection device; and
predicting a measurable crop outcome using the trained predictive model in response to the collected third data set and the collected fourth data set; wherein the predictive output comprises a mapping one of the group comprised of: a habitat suitability, a suggested varietal to plant, an expected quality of a crop, or a potential plantable crop varietal.