Shading Topography Imaging for Robotic Carton Interface Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vision systems for robotic unloading struggle to accurately determine the location of cartons in a stacked wall within a trailer, especially when there are minimal depth and visual variations between cartons, and require extensive calibration due to environmental lighting changes and specular reflections.
Innovation Solution
A vision system utilizing two light sources with distinct lenses positioned at different angles to create a shading topography image, which combines information from multiple images captured under varying illumination schemes, allowing the system to determine cargo position without requiring calibration for each instance and independent of depth variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional vision systems are used to detect carton positions, then the system can operate with standard imaging equipment, but the system fails to accurately determine carton locations when depth variations are minimal and requires extensive calibration for each lighting condition
Solution Approach 1:
The patent transitions from standard 2D imaging to shading topography imaging that captures depth information through light shadow patterns. By using multiple light sources at different angles and processing the differential shading information, the system creates a topographic map of the carton wall that reveals depth variations invisible to conventional cameras, enabling accurate carton interface detection without extensive calibration
Solution Approach 2:
The system changes the illumination parameters by using multiple light sources with different trajectories (upper and lower directional lighting) instead of single-source illumination. This parameter change in lighting geometry creates distinct shading patterns that highlight depth variations and carton interfaces, allowing the vision system to detect positions accurately across varying environmental conditions without recalibration
2Reliability
If multiple light sources with different trajectories are used to illuminate the cargo, then the system can detect carton interfaces independently of depth variations, but the imaging system becomes more complex requiring multiple light sources and sophisticated image processing
Solution Approach 1:
The illumination system is segmented into multiple independent light sources with distinct trajectories (upper and lower lights). Each light source captures different aspects of the cargo surface shading, and the vision control system processes these segmented images separately before blending them into a composite shading topography image. This segmentation allows the system to overcome the complexity by treating each light source as an independent detection channel
Solution Approach 2:
The patent merges multiple images captured under different illumination schemes into a single blended composite shading topography image. By combining the information from upper and lower light trajectories, the system creates a comprehensive view that eliminates depth variation dependencies and provides reliable carton interface detection, making the complex multi-light system function as a unified reliable detector
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables precise detection of carton interfaces and positions within a stacked wall, improving robotic unloading efficiency by eliminating the need for extensive calibration and handling variations in environmental lighting, while maintaining accuracy across different scenarios.
Implementation Method 1
two light sources spaced apart from each other... capture images of cargo... capture images of cargo under illumination by at least one of the at least two light sources
Data Source
AI summary
Vision systems for robotic assemblies for handling cargo, for example, unloading cargo from a trailer, can determine the position of cargo based on shading topography. Shading topography imaging can be performed by using light sources arranged at different positions relative to the image capture device(s).