Agricultural Sensor Localization for Aberration-Filtered Geospatial Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing agricultural systems face challenges in obtaining meaningful instantaneous sensor values due to noise and variability, leading to inaccuracies and errors in aggregated sensor data, which can be affected by aberrant spikes, compromising the accuracy of geospatial data correlation.

Innovation Solution

A system that aggregates sensor samples to generate localized sensor values by analyzing and correcting aberrant samples, ensuring more accurate representation of geographic locations, and generating control signals based on these corrected values.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If sensor samples are aggregated to obtain aggregated sensor values, then measurement noise is reduced, but aberrant spikes and inaccuracies are introduced into the geospatial data

Engineering Contradiction:
Improvesensor data accuracyVSAvoidgeospatial data accuracy
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent segments the aggregated sensor data by identifying individual sensor samples within the aggregated value and associating each with specific geographic locations. This allows the system to divide and conquer the problem by treating each sample individually rather than as a bulk aggregate, enabling precise localization and aberration detection.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts aberrant sensor samples from the aggregated data by comparing individual sample values against threshold criteria. When a sample exceeds the threshold, it is identified as an aberration and excluded from the geospatial correlation, preventing it from compromising the overall data accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

2Speed

If instantaneous sensor values are used, then real-time responsiveness is improved, but noise and variability reduce data meaningfulness

Engineering Contradiction:
Improvedata processing speedVSAvoidsensor value accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent performs preliminary aggregation of sensor samples into aggregated values before geospatial correlation. This preliminary action reduces noise and variability by combining multiple instantaneous readings, creating a more meaningful baseline value that represents the sensor data more accurately.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements a feedback mechanism where individual sensor samples within the aggregated value are examined and compared against threshold criteria. This feedback loop allows the system to identify and correct aberrant samples, improving the overall accuracy of the geospatial data while maintaining real-time processing capabilities.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If aggregated sensor values are used for geospatial correlation, then data representativeness is improved, but aberrant samples compromise accuracy

Engineering Contradiction:
Improvegeospatial data representativenessVSAvoidcontrol signal accuracy
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent extracts and removes aberrant sensor samples from the aggregated data by comparing each sample against threshold criteria. This extraction process ensures that only accurate, representative samples are used for geospatial correlation, preventing aberrations from compromising control signal accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies local quality control by examining each individual sensor sample within the aggregated value and applying threshold-based filtering. This localized approach ensures that each sample meets quality standards before being included in the geospatial correlation, maintaining high data reliability.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP4367998B1Agricultural system and method of controlling such
Publication Date: 2025.08.20 DEERE & CO
  • EP4367998B1 patent drawingFigure 1A
  • EP4367998B1 patent drawingFigure 1B
  • EP4367998B1 patent drawingFigure 1C

AI summary

An agricultural system is disclosed. The agricultural system comprising: a sensor (194) sensing a characteristic of an agricultural operation performed by an agricultural machine and generating a sensor signal responsive to the sensed characteristic; a signal processor (196) that aggregates a first plurality of samples of the sensor signal to obtain a first aggregated signal value; a sample geospatial correlation system (217) identifying a first geographic location corresponding to the first aggregated signal value; a sample localization system (219) that generates a localized signal value based on a subset of the first plurality of samples, the sample geospatial correlation system (217) identifying a second geographic location corresponding to the localized signal value; and an action signal generator (230) that generates an action signal based on the localized signal value. Furthermore, a method of controlling an agricultural system is disclosed.