Multi-Camera Image Processing for Autonomous Agricultural Machines

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current autonomous agricultural work machines are costly due to the need for multiple computing units for each camera, increasing manufacturing costs and installation space requirements.

Innovation Solution

Implementing a single computing unit to evaluate images from multiple cameras, allowing for cost-effective image processing and display, with the option to adjust machine components and use predefined functions or artificial intelligence for obstacle detection and field navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If multiple computing units are used for each camera to enable autonomous operation and image display, then the machine's autonomous capability and image processing quality are improved, but the manufacturing cost and device complexity increase significantly

Engineering Contradiction:
Improveautonomous capabilityVSAvoiddevice complexity
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The patent merges the computing functions for multiple cameras into a single shared computing unit. Instead of having separate computing units for each camera, one computing unit processes images from all cameras, reducing the number of computing components while maintaining autonomous operation capability. This combining approach directly reduces device complexity and manufacturing cost.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The single computing unit is designed to perform multiple functions: it processes images from multiple different cameras, enables autonomous navigation, displays images on monitors, and supports various operating modes. This multi-functional design allows one computing unit to replace what would traditionally require multiple specialized computing units.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Measurement precision

If multiple computing units are used for each camera, then the image processing capability is improved, but the manufacturing cost increases

Engineering Contradiction:
Improveimage processing capabilityVSAvoidmanufacturing cost
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The computing functions are merged into a single unit that handles all camera inputs. This consolidation reduces the total number of computing components needed, directly lowering manufacturing costs while maintaining the ability to process images from multiple cameras simultaneously or sequentially.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The single computing unit is designed with universal processing capabilities that can handle images from any of the multiple cameras. It performs obstacle detection, navigation processing, and image display functions regardless of which camera is capturing the image, providing cost-effective multi-functional processing.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If multiple computing units are used for each camera, then the image evaluation quality is improved, but the installation space requirements increase

Engineering Contradiction:
Improveimage evaluation qualityVSAvoidinstallation space
Core Design Contradiction:
Measurement precisionVSArea of stationary object

Solution Approach 1:

By combining the computing functions into a single unit rather than having separate units for each camera, the physical space required for housing these computing components is significantly reduced. The single computing unit can be installed in one location within the vehicle, reducing the overall installation footprint.

Inventive Principle:
Principle #5Merging (Combining)

4Ease of manufacture

If a single computing unit is used for all cameras, then the manufacturing cost is reduced, but the image processing load on the single unit increases

Engineering Contradiction:
Improvemanufacturing costVSAvoidimage processing load
Core Design Contradiction:
Ease of manufactureVSPower

Solution Approach 1:

The computing unit processes images from multiple cameras in a segmented manner, handling one camera's image data at a time or in sequential batches rather than attempting to process all camera inputs simultaneously. This temporal segmentation reduces the peak processing load on the single computing unit while maintaining the ability to evaluate all camera feeds.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4449842A1Agricultural machine
Publication Date: 2024.10.23 CLAAS E SYSTEMS GMBH
  • EP4449842A1 patent drawingFigure 1
  • EP4449842A1 patent drawingFigure 2
  • EP4449842A1 patent drawing

AI summary

The present invention relates to an agricultural machine (1) for use in an agricultural field (2) with several cameras (3, 4, 5, 6, 7, 8, 9) for generating multiple images of several surrounding areas of the agricultural machine (1). The present invention is based on the general idea that only one processing unit (13) evaluates the images transmitted by the cameras (3, 4, 5, 6, 7, 8, 9).