BR-102025006426-A2 - COMPUTER-IMPLEMENTED METHOD, AND SYSTEM

BR102025006426A2BR 102025006426 A2BR102025006426 A2BR 102025006426A2BR-102025006426-A2

Abstract

Detection and classification of internal plant parts are provided for work machines, e.g., agricultural machinery, control. The first input signals (e.g., camera images) correspond to a field of view including plant parts within a traversed work area, where the external attributes of the plant parts are identified based on the first input signals received. The second input signals correspond to penetrating radiation (e.g., non-ionizing) directed at the plant parts, where at least one internal attribute of the plant part is determined in relation to the plant parts, based on the second input signals received. Exemplary attributes may correspond to identified interior silhouettes, layers, boundaries, voids, shapes, and the like. Action signals are generated corresponding to an operation in the work area, based on at least one determined attribute of the internal plant part. Action signals may be generated for display, crop care control, harvesting machine control, harvesting logistics control, and the like.

Inventors

Mahesh Somarowthu
NOEL W. ANDERSON
TIMOTHY A. WILCOX

Assignees

DEERE & COMPANY

Dates

Publication Date: 20260310
Application Date: 20250331
Priority Date: 20240829

Claims (12)

1. Computer-implemented method (300), characterized in that it comprises: receiving (310) one or more first input signals corresponding to a field of view comprising one or more plant parts in a work area being traversed by a work machine (102), wherein the field of view is associated with one or more external attribute sensors (216) associated with the work machine; identifying (312, 320) an external attribute of at least one of the one or more plant parts based on the one or more first input signals received; receiving (310) one or more second input signals corresponding to penetrating radiation directed to at least one of the one or more plant parts, wherein the penetrating radiation is received by one or more internal attribute sensors associated with the work machine; determining (311, 320) at least one internal plant part attribute with respect to at least one of the one or more plant parts, based on at least one or more second input signals received; and generate action signals (330) corresponding to an operation in the work area, based on at least one attribute of a determined internal part of the plant.
2. A computer-implemented method according to claim 1, characterized in that at least one of the external attribute sensors and at least one of the one or more internal attribute sensors are the same device.
3. A computer-implemented method according to claim 1, characterized in that it comprises: training one or more first models over time to correlate external attributes identified based on the first input signals with the respective plant part types; and training one or more second models over time to correlate penetrating radiation characteristics associated with the respective plant part types with internal plant part attributes of the respective plant part types; wherein, for a current set of one or more first input signals and a corresponding set of one or more second input signals, at least one attribute of the internal plant part is determined with respect to at least one of the one or more plant parts by reference to a selection of the one or more first models trained to identify a first plant part type and by reference to a selection of the one or more second models trained to determine at least one attribute of the internal plant part.
4. A computer-implemented method according to claim 1, characterized in that it comprises generating action signals based on a defined aggregate of attributes of internal plant parts with respect to a defined period of time and/or with respect to a defined distance traveled by the work machine and/or with respect to the work area being traversed.
5. A computer-implemented method according to claim 1, characterized in that one or more external attribute sensors and one or more internal attribute sensors are mounted on a first working machine, and action signals corresponding to the operation are generated by a second working machine in the work area.
6. A computer-implemented method according to claim 5, characterized in that the action signals generated by the first working machine are used to map the determined attribute of at least one internal part of the plant to a location of the respective plant part in a selectively retrievable data structure, for each of the one or more plant parts having a determined internal plant part attribute for it.
7. A computer-implemented method according to claim 6, characterized in that the operation by the second working machine is controlled based on attributes mapped to the selectively retrievable data structure.
8. A computer-implemented method according to claim 1, characterized in that the operation is performed in the work area by the working machine and controlled based on the action signals generated.
9. A computer-implemented method according to claim 1, characterized in that the penetrating radiation comprises non-ionizing radiation generated by at least one emitter associated with the working machine.
10. System (100), characterized in that it comprises: one or more external attribute sensors (216) associated with a work machine (102) and configured to generate one or more first input signals (310) corresponding to a field of view of the one or more sensors and comprising one or more plant parts in a work area being traversed by the work machine; one or more internal attribute sensors (214) associated with the work machine and configured to generate one or more second input signals (310) corresponding to penetrating radiation directed to at least one of the one or more plant parts and received by the one or more second sensors; and one or more processors (202, 204, 206, 220) communicatively linked to one or more first sensors and to one or more second sensors to receive their respective input signals, and configured to direct the execution of steps in a method as defined in any of claims 1 to 9.
11. System according to claim 10, characterized in that one or more internal attribute sensors comprise at least one paired emitter and detector mounted on at least one side of the working machine and configured for reflection and backscatter measurements with respect to emitted penetrating radiation.
12. System according to claim 10, characterized in that one or more internal attribute sensors comprise at least one paired emitter and detector, arranged so that parts of the plant pass between the respective emitter and detector pairing for attenuation measurements with respect to the emitted penetrating radiation.

Description

DESCRIPTION FIELD [001] The present description refers, in general, to work machines with associated work implements, for example, towed or otherwise attached to self-propelled work vehicles, and more particularly to a method and system for controlling or otherwise generating action signals with respect to the operation of such work implements based on internal characteristics of plant parts detected and/or estimated within a work area being traversed by the work machine. FUNDAMENTALS [002] A work area may, for example, represent a field for growing a crop or other vegetation, or another type of area, including land to be worked by an implement or other tool associated with a work machine. [003] External plant attributes (also referred to here as external characteristics) have been used in some conventional systems and methods to control agricultural machinery and provide data to guide subsequent agricultural practices. Examples include, but are not limited to, the Normalized Difference Vegetation Index (NDVI) for nitrogen applications, biomass (e.g., height x population, leaf area index) for combine harvester feed rate control, stem diameter for floor plate clearance adjustments, and the like. [004] Internal plant attributes, or internal characteristics, particularly those related to seeds or pests obscured by husks (e.g., corn cobs), pods (e.g., soybeans), heads (e.g., small grains) and shells (e.g., nuts), or those related to fleshy bodies under husks, such as fruits (e.g., apples, grapes) or tubers (e.g., potatoes) have not been widely used for the control of self-propelled or mobile agricultural machinery. [005] X-ray, gamma-ray, positron emission tomography (PET), and magnetic resonance imaging (MRI) sensors have been used in conventional systems and methods for detecting the interiors of plant parts in laboratories and food processing plants. These technologies can be highly effective in the right context, but are not easily transferred to agricultural field environments due to the sensor size, cost, and regulations required (e.g., X-rays and gamma rays are highly regulated due to ionizing radiation and radioactive sources). The detected internal attributes have also not been actionable in conventional applications, either for real-time control or logistics for current or subsequent stations, at least due to the lack of in situ detection/estimation and/or georeferencing of internal attributes for individual or collective groups of plant parts. BRIEF SUMMARY [006] Various system modalities and/or methods as described herein are provided to address some or all of the problems referenced above with respect to the detection of internal plant parts and corresponding use for mobile agricultural machine control, at least in part using penetrating signals (e.g., non-ionizing) that can be generated and received with accessible and machine-mountable components and processed to generate action signals for display, for crop care control, for harvesting machine control, for harvest logistics control and the like. [007] In a particular and exemplary embodiment, a computer-implemented method, as described herein, may comprise: receiving one or more first input signals corresponding to a field of view comprising one or more plant parts in a work area being traversed by a work machine, wherein the field of view is associated with one or more external attribute sensors associated with the work machine; identifying an external attribute (e.g., contour) of at least one of the one or more plant parts based on the one or more first input signals received; receiving one or more second input signals corresponding to penetrating radiation directed to at least one of the one or more plant parts, wherein the penetrating radiation is received by one or more internal attribute sensors associated with the work machine; determining at least one internal plant part attribute with respect to at least one of the one or more plant parts, based on at least one or more second input signals received; and generating action signals corresponding to an operation in the work area, based on at least one determined internal plant part attribute. [008] In an exemplary aspect according to the embodiment referenced above, at least one of the one or more external attribute sensors and at least one of the internal attribute sensors may be the same device. As an example, backscattering can be used to identify the soybean seed and then the transmitted radiation can provide information about the interior of the seed. In another example, a first received signal could be from the outside of the seed, identifying the seed, and then a reflected signal determined by internal characteristics. [009] In another exemplary aspect according to the above-referenced embodiment, the method may comprise training one or more first models over time to correlate external attributes identified based on the first input signals with respective plant part types, and traini