patent.fyi
Applications  /  US 2026/0354296
United States Patent Application Publication

US 2026/0354296 A1

Published
Nov 12, 2026
Filed
Dec 5, 2023
Earliest date on the face
Dec 23, 2022
Classification
A01B 79/005
First published here
Nov 18, 2026

Generated automatically. Not reviewed by a person. Not legal advice or an opinion.

Section 1 · machine-generated

What if?

Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.

1.1Physical measurements guide work

What ifWhat if the predictor used only soil moisture, slope, and bare-ground fraction to estimate erosion after each tillage pass?
Claim language“including biological diversity of the ecosystem”

Capacitance probes measure soil moisture, an elevation grid supplies slope, and overhead images supply bare-ground fraction. A runoff model uses implement type and tillage depth to estimate sediment loss for each candidate pass under a fixed storm scenario.

1.1Field trials measure outcomes

What ifWhat if the system measured ecosystem conditions after pilot work on test strips and used those observations to recommend a treatment?
Claim language“predicts a state of an ecosystem after a work”

Replicated strips receive mowing treatments at different cutting heights, and untreated strips provide a baseline. Quadrats record plant species counts and ground cover four weeks later, and the system ranks the tested treatments by their measured differences from the baseline.

1.2Machine settings govern selection

What ifWhat if the evaluator checked proposed machine settings against equipment and field limits?
Claim language“evaluates a prediction state”

The evaluator checks candidate tillage depths against the implement travel limit and the measured thickness of the topsoil. It selects the widest pass that meets both limits, then a predictor displays the estimated ground cover for that pass.

1.2Sensor thresholds trigger work

What ifWhat if the evaluator scored current field measurements to decide which zone needed work first?
Claim language“evaluates a prediction state”

Tensiometers report current soil water tension for each irrigation zone. The controller selects the zone with the largest measured excess above its tension setpoint and assigns a pulse duration from its measured water deficit and emitter flow rate.

1.3Calendar sets mowing proposals

What ifWhat if a fixed field calendar determined the proposed mowing pass and its settings?
Claim language“according to an evaluation result of the prediction state”

A table keyed by field and week supplies the scheduled mowing zone, cutting height, and travel speed. After the proposal is set, a separate display shows the estimated post-mowing ground cover and its rating.

Section 2 · machine-generated

Variations

Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.

Other field of use

Vegetation management at a solar facility

A vegetation management controller divides a solar facility into strips beneath panel rows and along service lanes. It uses plant species counts and cover by species as the starting state, together with shade duration and soil moisture, to predict vegetation cover after mowing or cuttings removal at a specified height. It scores the predicted state against minimum native plant richness and maximum vegetation height at access points, then proposes a work type, strip identifier, and cutting height.

Component substitution

Grid simulation of meadow recovery

A cellular automaton serves as the prediction unit and represents a meadow as a grid with plant species, vegetation height, and bare ground fraction recorded for each cell. The simulator applies the specified operation within the work footprint by changing vegetation height for mowing or increasing bare ground fraction for scarification. Each weekly step updates plant cover using local regrowth rates and seed input from neighboring cells. The evaluation unit scores the predicted meadow by native plant richness and bare ground fraction, and the determination unit proposes the operation and settings when the score meets a fixed threshold.

Parameter or scale variation

Two season scheduling for streambank mowing

A controller divides a 2 km streambank into 40 segments of 50 m and records native plant richness and flowering cover for each segment. For a mowing operation specified by date and cutting height, the controller predicts insect-pollinated plant cover at 30, 180, and 365 days after the cut. An evaluation unit checks whether at least 25 percent of the bank retains flowering cover during the first 30 days and whether native richness returns to its starting level within 365 days. A scheduling unit proposes the cut date and height for each accepted segment and groups the segments into spring and autumn work periods.

Combination with existing systems

Habitat work orders for drainage maintenance

A habitat planning module connects to a geographic asset management system through the identifiers assigned to ditch reaches. It reads aquatic plant richness, vegetation cover, and channel geometry, then predicts residual habitat cover and open channel width after a specified length and bank side are cleared. The evaluator checks both minimum habitat cover and minimum open width, and the module writes the accepted clearing length, bank side, and reach identifier to a draft maintenance work order.

Selective thinning with harvester control maps

A forest planning module joins tree species and stem locations from a stand inventory with the map used by a forestry harvester. For a thinning prescription defined by minimum stem diameter and removal density, the module subtracts the designated stems and predicts residual canopy cover and tree species distribution. It scores the predicted stand by retained tree species richness and the area of connected canopy gaps. When the score meets the acceptance threshold, the module exports a georeferenced stem selection layer for display on the harvester terminal.

Automated or data-driven variant

Camera updates for grazing proposals

Fixed cameras capture daily images of pasture quadrats, and an image classifier estimates cover by plant group for each quadrat. Software combines the estimates with the latest field counts of plant species, then runs a regrowth model using the proposed stocking density and grazing duration. An evaluator checks predicted residual cover and flowering forb cover against fixed thresholds, and the system issues a grazing proposal with the paddock, stocking density, and duration when both thresholds are met.

Specific version of a part

Regression predicts plant richness

A count regression model takes current quadrat species richness, vegetation cover, soil moisture, implement type, and cutting height as inputs. It estimates plant species richness per square metre six weeks after each candidate mowing pass.

Grid model predicts habitat

A raster model assigns each five-metre cell its measured plant species richness, canopy cover, and soil moisture. Transition tables indexed by brush-cutting equipment type and cutting height update those cell values to estimate conditions 30 days after treatment.

Thresholds score predicted outcomes

A rules engine checks whether predicted native plant richness reaches eight species per square metre and whether exposed soil stays below 15 percent. For each threshold it computes a normalized margin, then assigns the smaller margin as the evaluation score.

Ranking selects work settings

A selector ranks candidate work records by their ecosystem evaluation score and breaks ties using the shorter scheduled machine time. It copies the selected record’s mower type, cutting height, and travel speed into the proposal.

Section 3 · quoted from USPTO records

Other work

Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.

Document AUS 2009/0307670 A1
Published
Dec 10, 2009
Filed
Jun 7, 2008
Class
G06F 9/44
Document BUS 10,878,289 B2
Published
Dec 29, 2020
Filed
Nov 22, 2019
Class
G06K 9/6257
Claim 1
Document A
Document B
1.1a prediction unit that predicts a state of an ecosystem after a work in a case where a work indicated by work information including a type of the work by a work device and a work parameter is performed on the ecosystem by using state information regarding a state of the ecosystem including biological diversity of the ecosystem;
For example, the simulation can be performed by having the inference engine 6 rerun the data evaluators 4C based on the historical parameter data previously collected by the collection mechanisms 4B, but assuming a modified system that has been reconfigured according to one (or more) of the reconfiguration recommendations (which can be fed back to the data evaluators).
A · US 2009/0307670 A1 · ¶ 35
The social-ecological model is a software algorithm that takes as input a parameterised data set that models the state of the environment at a particular point in time, and outputs a similarly parameterised data set that reflects the state of the environment in response to the action having been carried out.
B · US 10,878,289 B2 · ¶ 65
1.2an evaluation unit that evaluates a prediction state that is the predicted state of the ecosystem; and
Scenario evaluation is performed based on simulation of the reconfiguration recommendation using the data evaluation result to determine efficacy of the scenario.
A · US 2009/0307670 A1 · abstract
1.3a determination unit that determines proposal work information that is the work information to be proposed according to an evaluation result of the prediction state.
The knowledge base is further consulted to determine a reconfiguration recommendation based on the data evaluation result.
A · US 2009/0307670 A1 · ¶ 10
Based on the results of the simulation, a determination is made as to an appropriate action (or course of action) to perform in the environment (step S104).
B · US 10,878,289 B2 · ¶ 65

An empty cell means only that no passage was reproduced there.

Statements in these documents
An automated mechanism that provides recommendations for system optimization in an efficient manner without the aforementioned limitations would be desirable.
A · US 2009/0307670 A1 · ¶ 9
Scenario evaluation is performed based on simulation of the reconfiguration recommendation to determine the efficacy of the scenario without having to actually implement the currently unavailable system features or parameters.
A · US 2009/0307670 A1 · ¶ 10
By modelling the response of actors and systems to different stimuli (such as user-input commands, or computer-generated commands), it is possible to simulate the outcome of different courses of action, and so help to determine an appropriate strategy when confronted with the same choices in the real world.
B · US 10,878,289 B2 · ¶ 2
It follows that there is a need to develop tools for enhancing the range and/or accuracy of computer simulations of real-world scenarios.
B · US 10,878,289 B2 · ¶ 6
Cite aspatent.fyi, US 2026/0354296 A1, claim 1, first published 2026-11-18T21:18:11-05:00, https://patent.fyi/US20260354296/, SHA-256 5791dc0f808338b519fcd23eebefea6f7f7b5eda695f4731519d4030c737b374
First published
Permanent addresshttps://patent.fyi/US20260354296/
Content fingerprint (SHA-256)5791dc0f808338b519fcd23eebefea6f7f7b5eda695f4731519d4030c737b374
Methodv2026-11-18.2 · automated, not reviewed by a person
Correctionscorrections@patent.fyi