Terrain Image Analysis for Mobile Work Machine Path Hazards

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Mobile agricultural machines, such as combine harvesters, face operational challenges due to varying terrain conditions like standing water and mud, which can cause the machines to become stuck, and these conditions are often obscured from the operator's view, leading to adverse performance.

Innovation Solution

A terrain image analysis system that captures images of the path ahead using cameras or UAVs, performs spectral analysis, and compares them to a baseline model to detect changes in terrain conditions, generating warnings and control signals to adjust the machine's operation, such as speed and steering, to avoid adverse conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the machine traverses over terrain with varying soil conditions, then the machine can operate in diverse agricultural environments, but the machine may encounter standing water or mud that causes it to become stuck

Engineering Contradiction:
Improveability to operate in diverse terrain conditionsVSAvoidrisk of machine becoming stuck
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The system performs preliminary detection of terrain conditions using cameras and spectral analysis to identify standing water or mud before the machine reaches hazardous areas. This advance warning allows the operator to take preventive action, such as adjusting speed or steering, to avoid becoming stuck while maintaining the ability to operate in diverse agricultural environments

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If the operator relies on direct visual observation of terrain, then the operator can perceive terrain conditions, but adverse conditions like standing water are often obscured from view

Engineering Contradiction:
Improveoperator's ability to perceive terrainVSAvoidterrain conditions obscured from operator's view
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The system replaces the operator's direct visual observation with an automated optical detection system using cameras and spectral analysis. This electronic sensing system can detect terrain conditions such as standing water and mud that are obscured from the operator's view, converting optical information into actionable warnings without requiring the operator to directly see the hazardous conditions

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If the machine operates at high speed to improve productivity, then more area can be harvested per unit time, but the machine is more likely to become stuck in adverse terrain conditions

Engineering Contradiction:
Improveharvesting area per unit timeVSAvoidlikelihood of becoming stuck
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system provides real-time feedback about terrain conditions to the operator through warnings and alerts. This feedback loop allows the operator to maintain high speeds in safe conditions while automatically reducing speed or adjusting operation when hazardous terrain is detected, thereby maintaining productivity while reducing the likelihood of becoming stuck

Inventive Principle:
Principle #23Feedback

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The system effectively alerts operators to potential hazards and adjusts the machine's operation to prevent getting stuck, improving performance and safety by providing real-time terrain awareness.

Implementation Method 1

performs spectral analysis, and compares them to a baseline model to detect changes in terrain conditions

Methodology Applied
Scientific EffectSpectral analysis: Absorption Spectroscopy

Data Source

PatentEP3739510B1Mobile work machine control system with terrain image analysis
Publication Date: 2023.06.14 DEERE & CO
  • EP3739510B1 patent drawingFigure 1
  • EP3739510B1 patent drawingFigure 2
  • EP3739510B1 patent drawingFigure 3

AI summary

A method of controlling a mobile work machine includes receiving an image of spectral response at an area of terrain corresponding to a path of the mobile work machine, generating an image distance metric based on a distance between the spectral response and a base spectral response model corresponding to the terrain, comparing the image distance metric to a distance threshold, and controlling a controllable subsystem of the mobile work machine based on the comparison.