Robot Picking Curved Objects Gray Gradient Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face difficulties in detecting and picking up workpieces with curved shapes due to lack of detectable textures and contours, and require complex equipment and extensive processing, making them impractical for use with robots.
Innovation Solution
An apparatus using a robot with a gray gradient distribution model and edge information model to detect and pick up workpieces with slight texture and unclear contours, incorporating a robot moving mechanism to accurately position the robot for picking up, and a floating mechanism to accommodate position errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If workpieces with curved shapes are detected using traditional contour and texture methods, then detection accuracy improves for polyhedral workpieces, but detection becomes impossible for workpieces without detectable contours or textures
Solution Approach 1:
The invention changes the detection parameter from contour and texture features to gray gradient distribution features. By extracting gray gradient information in multiple directions (horizontal, vertical, diagonal) and comparing it with pre-stored models, the system can detect curved shape workpieces that lack traditional detectable features like contours or textures.
Solution Approach 2:
The invention creates gray gradient distribution models that represent the characteristic gradient patterns of workpieces. These models serve as templates for detection, allowing the system to identify workpieces by matching their gray gradient distributions against the stored models, enabling detection of curved shapes without relying on their physical contours or textures.
2Adaptability or versatility
If complex equipment like laser slit light scanning is used to obtain three-dimensional shape data, then detection capability for curved workpieces improves, but device complexity and cost increase significantly
Solution Approach 1:
The invention extracts only the essential gray gradient distribution information from the image data, discarding the need for complex three-dimensional scanning equipment. By focusing on the gradient characteristics in the two-dimensional image plane, the system achieves curved workpiece detection using simple imaging devices rather than sophisticated laser scanning systems.
Solution Approach 2:
The invention replaces complex mechanical scanning systems (laser slit light scanners) with a simpler optical imaging system combined with image processing. Instead of using physical laser scanning to obtain 3D data, the system uses gray gradient analysis on standard images to achieve detection of curved workpieces.
3Measurement precision
If extensive calculating processing is performed to calculate three-dimensional data from scanned images, then detection accuracy improves, but processing time increases making the system impractical
Solution Approach 1:
The invention performs preliminary action by pre-calculating and storing gray gradient distribution models for different workpiece types before actual detection. During operation, the system only needs to extract gray gradients from the current image and compare them with the pre-stored models, avoiding the need for extensive real-time calculations and enabling rapid detection.
4Measurement precision
If workpieces are detected using methods requiring complete separation from background, then detection precision improves for isolated objects, but detection fails when multiple workpieces are superposed
Solution Approach 1:
The invention applies local quality analysis by examining gray gradient characteristics at different locations and directions within the image. By analyzing the local gray gradient distribution patterns and comparing them with stored models, the system can identify individual workpieces even when they are superposed, as each workpiece contributes its characteristic gradient pattern to the overall image.
Data Source
Figure 1
Figure 2a~2b
Figure 3
AI summary
An apparatus for picking up objects including a robot (1) for picking up an object (6), at least one part of the object having a curved shape, having a storing means (30) for storing a gray gradient distribution model of the object (6), a recognizing means (2) for recognizing a gray image of the object (6), a gradient extracting means (34) for extracting a gray gradient distribution from the gray image recognized by the recognizing means (2), an object detecting means (26) for detecting a position or position posture of the object (6) in the gray image in accordance with the gray gradient distribution extracted by the gradient extracting means (34) and the gray gradient distribution model stored by the storing means (30), a detection information converting means (37) for converting information of the position or position posture detected by the object detecting means (36) into information of position or position posture in a coordinate system regarding the robot (1); and a robot moving means (1a) for moving the robot (1) to the position or position posture converted by the detection information converting means (37) to pick up the object (6). Thus, the object (6) having a curved shape can be detected and picked up in a reasonably short period of time.