US 12,149,582 B2
Dynamic allocation of platform-independent machine learning state machines between edge-based and cloud-based computing resources
Yueqi Li, San Jose, CA (US); and Alexander Ngai, Irvine, CA (US)
Assigned to Deere & Company, Moline, IL (US)
Filed by Deere & Company, Moline, IL (US)
Filed on Dec. 1, 2021, as Appl. No. 17/540,015.
Prior Publication US 2023/0171303 A1, Jun. 1, 2023
Int. Cl. G06F 15/16 (2006.01); G06F 9/48 (2006.01); G06F 9/50 (2006.01); G06Q 50/02 (2012.01); H04L 67/10 (2022.01); H04L 67/12 (2022.01)
CPC H04L 67/10 (2013.01) [G06F 9/4806 (2013.01); G06F 9/4881 (2013.01); G06F 9/5027 (2013.01); G06Q 50/02 (2013.01); H04L 67/12 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method implemented using one or more processors, comprising:
causing one or more graphical user interfaces (GUIs) to be rendered on a display, wherein each GUI of the one or more GUIs includes a working canvas that is usable to create, edit, manipulate, or otherwise act upon a plurality of graphical nodes corresponding to a plurality of platform-independent logical routines, wherein the plurality of graphical nodes are movable on the working canvas relative to each other to connect an output of at least one of the graphical nodes to an input of another of the graphical nodes via one or more edges to define a platform independent agricultural state machine, and wherein one or more of the platform independent logical routines includes logical operations that process agricultural data that represents one or more plants using one or more phenotyping machine learning models;
identifying one or more edge computing resources available to a user for which the agricultural state machine is to be implemented;
determining one or more constraints imposed by the user on implementation of the agricultural state machine, wherein the one or more constraints include one or more latency thresholds for obtaining phenotypic predictions about plants generated based on processing images of the plants using one or more of the platform-independent logical routines; and
based on the one or more edge computing resources and the one or more constraints,
dynamically allocating, to one or more of the edge computing resources, logical operations of one or more of the platform-independent logical routines of the agricultural state machine that will satisfy the one or more constraints when executed on one or more of the edge computing resources; and
dynamically allocating, to a cloud computing resource in network communication with the one or more edge computing resources, logical operations of one or more other platform-independent logical routines of the agricultural state machine that will not satisfy the one or more constraints when executed on one or more of the edge computing resources.