Work Vehicle Obstacle Classification for Selective Avoidance Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional work vehicles inefficiently execute avoidance maneuvers due to determining obstacles based solely on distance, leading to decreased work efficiency when encountering objects that do not require avoidance, such as reaping or harvesting targets.

Innovation Solution

A vehicle control method that acquires distance and image information to determine the type of a detection target, enabling appropriate countermeasure processing based on the target's type and distance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the work vehicle determines obstacles based solely on distance threshold, then the vehicle can quickly identify potential hazards and execute avoidance maneuvers, but the vehicle incorrectly treats work target objects as obstacles, causing unnecessary avoidance motions and reduced work efficiency

Engineering Contradiction:
Improveobstacle detection accuracyVSAvoidwork efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The detection targets are segmented into different categories (obstacles vs. work target objects) based on image recognition analysis. The imaging unit captures images of detection targets, and the control unit analyzes these images to determine whether each target is an obstacle or a work target object, enabling differentiated response actions for each category.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

An image recognition analysis mechanism serves as an intermediary between distance detection and avoidance control. The imaging unit and control unit work together to analyze images of detection targets, providing additional information that mediates the decision-making process to distinguish between obstacles and work target objects before executing avoidance maneuvers.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If the work vehicle executes avoidance motion for all detection targets within threshold distance, then collision avoidance is ensured, but unnecessary avoidance motions occur for work target objects, leading to time loss and reduced productivity

Engineering Contradiction:
Improvecollision avoidanceVSAvoidtime loss due to unnecessary avoidance
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The control system dynamically adjusts its response based on the type of detection target. When a work target object is identified through image analysis, the system suppresses avoidance motion even if the distance is within the threshold. When an obstacle is identified, the system executes avoidance motion. This dynamic adjustment eliminates unnecessary avoidance actions while maintaining collision avoidance for real hazards.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the control parameter (avoidance motion execution) based on the classification of the detection target. Image recognition analysis provides additional parameters (target type classification) that modify the original distance-based control logic, enabling selective suppression of avoidance motion for work target objects while maintaining it for obstacles.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4488788B1Vehicle control method, vehicle control program, and vehicle control system
Publication Date: 2026.02.25 YANMAR HLDG CO LTD
  • EP4488788B1 patent drawingFigure 1
  • EP4488788B1 patent drawingFigure 2~3
  • EP4488788B1 patent drawingFigure 4~5

AI summary

[Problem] To provide a vehicle control method, a vehicle control program, and a vehicle control system that prevent a decrease in the work efficiency of a work vehicle, while also being capable of causing the work vehicle to execute suitable countermeasure processing with respect to an obstacle. [Solution] An acquisition processing unit 511 acquires, when a work vehicle 10 performs autonomous travel according to a target route, distance information relating to a distance to a detection target that is detected by an obstacle sensor 54, and image information relating to a capture image of the detection target that is captured by a camera 53. A determination processing unit 512 determines a type of the detection target based on the image information. A countermeasure processing unit 112 causes the work vehicle 10 to execute countermeasure processing according to the type of the detection target and the distance information.