Soft Land Boundaries Using Variable Confidence Intervals

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for generating boundaries for vehicle operation on plots of land are burdensome due to the use of tabular geospatial data, which is large, difficult to manage, and costly to process. Additionally, these methods struggle to account for uncertainties in boundary locations, particularly in areas with discrepancies between data sources and actual conditions.

Innovation Solution

The system generates soft boundaries, which are probabilistic boundaries created by combining existing boundaries from different sources with weighted confidences. This approach compresses tabular geospatial data to analyze plots of land and produce geospatial features that correspond to the land, allowing for the assignment of machine operations within generated subzones.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If tabular geospatial data is used to generate boundaries, then boundary generation is possible, but data management becomes burdensome and costly

Engineering Contradiction:

Improveboundary generation capabilityVSAvoiddata management complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts only the essential boundary-defining features from comprehensive geospatial data, creating simplified boundary representations that retain operational reliability while eliminating redundant data management requirements

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the plot of land into subzones with clearly defined boundaries, allowing each boundary to be managed independently rather than as a monolithic data structure, reducing overall complexity

Inventive Principle:

Principle #1Segmentation

2Reliability

If traditional boundary methods are used, then boundaries can be defined, but uncertainties in boundary locations cannot be accounted for

Engineering Contradiction:

Improveboundary definition capabilityVSAvoidboundary uncertainty information

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent applies different confidence levels to different portions of boundaries, allowing high-certainty boundary segments to be defined precisely while accommodating uncertainty in other segments through probabilistic representations

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent transforms boundary representation from deterministic coordinates to probabilistic distributions, changing the parameter type from fixed position to confidence-based ranges that capture uncertainty

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If soft boundaries with probabilistic confidence are implemented, then uncertain boundary areas can be navigated, but boundary generation complexity increases

Engineering Contradiction:

Improveability to handle uncertain boundary areasVSAvoidboundary generation process complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements soft boundaries only where uncertainty exists, using traditional hard boundaries where confidence is high, thereby gaining adaptability in uncertain areas without unnecessarily complicating the entire boundary generation process

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS20250130582A1Methods and apparatus to determine soft boundaries through variable confidence intervals

Publication Date: 2025.04.24 DEERE & CO

AI summary

Systems, apparatus, articles of manufacture, and methods are disclosed to determine a boundary for vehicle operation in queried plot of land. An example apparatus includes circuitry to instantiate machine-readable instructions to: generate a first boundary based on a query for a boundary of a plot of land; compute a first probabilistic boundary for the first boundary based on an error of generation of the first boundary; compute a second probabilistic boundary for a second boundary based on an error of generation of the second boundary; and combine the first probabilistic boundary and a second probabilistic boundary to generate a soft boundary, the combination based on a first confidence score and a second confidence score.