Autonomous Work Machine Route Control for Sun Backlight Avoidance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous work machines face challenges in recognizing the external world due to complex image processing requirements and are unable to effectively avoid backlighting from the sun, which affects their ability to operate efficiently.

Innovation Solution

The system includes a camera-equipped autonomous work machine that uses wireless communication with a server to obtain and share sun information, allowing it to generate and implement routes that avoid backlighting by associating azimuth information with imaging date and time information and position data, thereby controlling its path to prevent backlighting.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If image processing is applied to recognize external world information, then the autonomous work machine can operate autonomously, but the image processing operation becomes more complicated under various capturing conditions

Engineering Contradiction:
Improveautonomous operation capabilityVSAvoidimage processing complexity
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The system performs preliminary action by obtaining sun position information (azimuth and altitude) in advance based on date and time, and pre-determining backlighting risk for each track segment before the autonomous work machine actually travels. This allows the complex backlighting avoidance calculations to be done beforehand, simplifying real-time decision-making.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary information layer (sun position data and pre-calculated backlighting risk flags) between the raw camera images and the image processing algorithm. Instead of directly processing complex images under varying lighting conditions, the system uses this intermediary data to predict and avoid backlighting scenarios, reducing the complexity of real-time image processing.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If the autonomous work machine travels through various tracks to work in the work area, then work coverage is improved, but backlighting from the sun may occur affecting image recognition

Engineering Contradiction:
Improvework coverageVSAvoidbacklighting effect
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The system divides the work area into multiple track segments and pre-calculates the backlighting risk for each segment based on the sun's position at different times and locations. This preliminary assessment allows the machine to select optimal tracks that avoid backlighting conditions while still achieving comprehensive work coverage.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements dynamic track selection by continuously updating the backlighting risk assessment as the autonomous work machine moves through the work area and as the sun's position changes throughout the day. The system dynamically adjusts the travel plan to avoid tracks with high backlighting risk, maintaining optimal image recognition conditions while preserving work coverage.

Inventive Principle:
Principle #15Dynamics

3Productivity

If the autonomous work machine operates during sunny periods for efficiency, then work speed is improved, but backlighting conditions deteriorate image quality

Engineering Contradiction:
Improvework speedVSAvoidimage recognition accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system obtains sun position information (azimuth and altitude angles) in advance based on the date and time of operation. By pre-calculating the backlighting risk for each track segment using this sun position data, the system can identify and select tracks that maintain good image quality even during sunny periods, thus preserving both work speed and recognition accuracy.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3874925B1Autonomous work machine, autonomous work machine control method, and program
Publication Date: 2023.06.14 HONDA MOTOR CO LTD
  • EP3874925B1 patent drawingFigure 1A~1B
  • EP3874925B1 patent drawingFigure 2
  • EP3874925B1 patent drawingFigure 3

AI summary

An autonomous work machine that includes a camera configured to capture an image of an external world, comprises an obtainment unit configured to obtain, based on date and time information, azimuth information of the sun, and imaging date and time information of the autonomous work machine, sun information that indicates one of a position and an azimuth of the sun in which backlighting can occur while the autonomous work machine is traveling in a work area, and a control unit configured to set, based on the sun information obtained by the obtainment unit, tracks by which the autonomous work machine will avoid backlighting.