Computer Vision Pose Estimation for Hinged Implement Positioning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing hinged vehicles face challenges in accurately positioning implements due to ground slope and undulations, relying on user skills and often requiring expensive positioning devices like GNSS and IMU.

Innovation Solution

A pose estimation control system using computer imaging to capture and analyze images of the implement, calculating its orientation and position, and adjusting the vehicle's position to compensate for shifts, thereby ensuring accurate implement positioning without the need for additional expensive sensors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If expensive positioning devices like GNSS and IMU are used on the implement, then positioning accuracy is improved, but device cost and complexity increase

Engineering Contradiction:
Improvepositioning accuracyVSAvoiddevice complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent uses computer vision to create a visual copy/image of the implement and analyzes this copy to determine position and orientation. Instead of placing sensors on the implement, the system captures images of the implement and processes these images to extract positioning information, effectively using a visual copy rather than physical sensors.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces mechanical positioning devices (GNSS receivers, IMU sensors) with an optical system. The computer vision system uses cameras to capture images and algorithms to process them, substituting the mechanical/electronic sensor-based approach with an optical imaging and computational approach.

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

2Measurement precision

If GNSS and IMU devices are installed on the implement, then positioning accuracy is improved, but the system cost increases

Engineering Contradiction:
Improvepositioning accuracyVSAvoidsystem cost
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The patent replaces expensive, sophisticated positioning hardware with more affordable computer vision components. The system uses standard cameras and processing algorithms that are less costly than GNSS receivers and IMU units, making the overall system more economically viable while achieving comparable or sufficient positioning accuracy.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The computer vision system serves multiple functions: it captures images for positioning analysis, determines both position and orientation of the implement, and can potentially be used for other monitoring purposes. This multi-functionality reduces the need for separate specialized devices, lowering overall system cost.

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

3Adaptability or versatility

If the implement is coupled by a hinge allowing positional independence, then adaptability to terrain is improved, but positioning accuracy deteriorates

Engineering Contradiction:
Improveadaptability to terrainVSAvoidpositioning accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent implements a feedback control system where the computer vision continuously monitors the implement's actual position and orientation, compares it to the desired position, and provides correction signals to the vehicle's steering system. This closed-loop feedback compensates for the positional independence introduced by the hinge connection, maintaining accuracy despite terrain variations.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The computer vision system acts as an intermediary between the hinged connection and the control system. It bridges the gap by visually measuring the implement's position and translating this information into corrective steering commands, mediating the effect of the hinge's positional independence on overall positioning accuracy.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4085420B1Pose estimation and applications using computer imaging
Publication Date: 2025.04.16 TRIMBLE INC
  • EP4085420B1 patent drawingFigure 1~2
  • EP4085420B1 patent drawingFigure 3
  • EP4085420B1 patent drawingFigure 4

AI summary

Embodiments describe a method for positioning a hinged vehicle including a primary part and a secondary part coupled to the primary part at a project site. The method includes receiving, from an image capturing device, digital image data representing one or more features of the secondary part; performing image analysis on the digital image data to identify positions of the one or more features of the secondary part; identifying an angle of at least a portion of the secondary part; calculating a current position of the secondary part based on the angle; calculating a positional difference between a correct position at the project site for the secondary part and a current position of the secondary part at the project site; and initiating a change in a position of the primary part to compensate for the positional difference and to position the secondary part on the correct position.