US 12,439,844 B2
Smart fertilizer delivery system
Aleksei Valentinovich Kramarenko, Chuvashiya (RU); Dmitry Mikhaylovich Mikhaylov, Moscow (RU); Alexey Sergeevich Pronichkin, Uliyanovskaya obl. (RU); Sergey Valerievich Bagrov, Moscow (RU); Natalya Mikhaylovna Zolotykh, Moscow (RU); and Ilman Shazhaev, Moscow (RU)
Assigned to PSIMOS, INC., New York, NY (US)
Filed by PSIMOS, Inc, New York, NY (US)
Filed on Jul. 20, 2021, as Appl. No. 17/380,953.
Claims priority of provisional application 63/053,837, filed on Jul. 20, 2020.
Prior Publication US 2022/0015287 A1, Jan. 20, 2022
Int. Cl. A01C 23/04 (2006.01); A01C 21/00 (2006.01); A01C 23/00 (2006.01); G01N 33/24 (2006.01); G05B 13/02 (2006.01); G06Q 50/02 (2024.01)
CPC A01C 23/042 (2013.01) [A01C 21/007 (2013.01); A01C 23/007 (2013.01); G01N 33/24 (2013.01); G05B 13/028 (2013.01); G06Q 50/02 (2013.01); G01N 33/245 (2024.05); G05B 2219/25297 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A system, comprising:
a memory configured to store data associated with an area to be fertilized, the data including information identifying a crop and soil analysis data associated with the area; and
a processor configured to:
receive sensor data associated with the area, the sensor data including information identifying a current weather condition and a current soil condition for the area;
automatically generate a proposed recipe of a fertilizer mixture including information identifying a combination of elements and a dosage of each of the elements, the proposed recipe being generated by a machine learning model associated with the area, the machine learning model being trained based on a combination of the sensor data and historical data associated with the area, executing with inputs including at least the current weather condition and the current soil condition, and generating a prediction specifically for the area including predictions for an optimized flow rate to deliver a mixture of the fertilizer mixture and water to the area and the crop by a delivery system;
determine whether the proposed recipe satisfies at least a first rule requiring that each of the elements of the proposed recipe are compatible;
determine whether the proposed recipe satisfies at least a second rule requiring that the proposed recipe complies with at least a first mixing rule that specifies whether the elements of the proposed recipe can be assimilated in the current soil condition; and
in an instance the proposed recipe satisfies both the at least first rule and the at least second rule, operate the delivery system, in accordance with the predictions for the optimized flow rate generated by the machine learning model, to automatically deliver the mixture of the fertilizer mixture and water to the area.