Robot Vision Coordinate Correction for Workpiece Height Variation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robot control systems face challenges in achieving high-accuracy positioning of workpieces due to parallax issues caused by using CCTV lenses, especially when imaging three-dimensional workpieces with varying heights, leading to deteriorated detection accuracy and workpiece handling precision.

Innovation Solution

A robot control system that utilizes a two-dimensional camera to image workpieces from a fixed height, processes the image to recognize positional coordinates in a two-dimensional vision coordinate system, and converts these coordinates into a three-dimensional world coordinate system by correcting for the workpiece's height dimension, allowing for precise robot arm positioning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If a CCTV lens is used in the camera to reduce cost, size, and weight, then the camera becomes more economical and compact, but parallax occurs causing deterioration in detection accuracy of positional coordinates

Engineering Contradiction:
Improvecamera costVSAvoiddetection accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent changes the parameter of lens type from conventional CCTV lens to telecentric lens. This parameter change eliminates parallax error by providing parallel light rays, thereby maintaining measurement precision while accepting higher cost. The telecentric lens ensures that the magnification remains constant regardless of object distance, solving the accuracy issue.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent uses a scale object (calibration target) that can be easily replaced or recalibrated. This allows the system to maintain accuracy by periodically updating the reference measurements, effectively compensating for any lens imperfections or drift over time without requiring expensive high-precision optics.

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

2Area of stationary object

If a camera is fixed above the arm movable range to provide wider visual field, then large workpieces can be imaged, but detection accuracy deteriorates in peripheral portions due to increased parallax

Engineering Contradiction:
Improvevisual fieldVSAvoiddetection accuracy
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent changes the optical parameter by using a telecentric lens instead of a conventional lens. This ensures that light rays are parallel to the optical axis, eliminating perspective distortion and parallax effects across the entire visual field. As a result, measurement accuracy is maintained from the center to the periphery of the image.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces the need for mechanical adjustment of camera position or multiple cameras with an optical solution (telecentric lens). This single lens provides both wide visual field coverage and uniform measurement accuracy across the entire field of view, eliminating the trade-off between visual field size and accuracy.

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

3Productivity

If multiple workpiece supply devices are provided or image processing is performed for multiple target regions, then supply capacity increases, but it becomes further difficult to secure accuracy when target region heights do not coincide

Engineering Contradiction:
Improvesupply capacityVSAvoidwork accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent uses a scale object as a reference copy for each workpiece or target region. By placing a known scale object at each supply device or target region, the system can independently calibrate and measure each region's coordinates without being affected by height differences. This copying approach allows multiple regions to be processed with consistent accuracy.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent segments the measurement space by introducing scale objects at each supply device or target region. Each scale object creates an independent reference frame, allowing the system to handle multiple workpieces at different heights independently. This segmentation enables parallel processing of multiple regions while maintaining individual accuracy for each segment.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3895855B1Robot control system and robot control method
Publication Date: 2026.04.15 FUJI CORP
  • EP3895855B1 patent drawingFigure 1
  • EP3895855B1 patent drawingFigure 2
  • EP3895855B1 patent drawingFigure 3

AI summary

An image of a workpiece (30) having a known height dimension and supplied to a work area (38) is captured by a two-dimensional camera (40) from a predetermined height position, the image is processed, and a position of the workpiece is recognized at coordinate values in a two-dimensional coordinate system (hereinafter, referred to as a "vision coordinate system") having a reference point of the image as an origin. The coordinate values in the vision coordinate system are converted into coordinate values in a world coordinate system that is a three-dimensional coordinate system of the robot (11), a target position of an arm of the robot is set at the coordinate values in the world coordinate system based on the converted position of the workpiece, and a position of the arm is controlled at the coordinate values in the world coordinate system. In this case, the coordinate values in the vision coordinate system recognized as the position of the workpiece through the image processing are corrected in accordance with a height dimension of the workpiece by taking into consideration that a size of the workpiece in the image captured by the camera changes depending on the height dimension of the workpiece, and the corrected coordinate values in the vision coordinate system are converted into the coordinate values in the world coordinate system.