Robotic Inkjet Camera Calibration for Large-Surface Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inkjet printing systems face challenges with large robots experiencing high dynamic loads and require precise camera calibration for low latency defect detection, particularly in large-scale applications like aircraft components, where nozzle outages and motion system errors are critical issues.
Innovation Solution
A method and system for calibrating cameras relative to the motion of a robotic inkjet system using calibration artifacts printed at known positions, enabling accurate localization and rectification of defects by determining camera intrinsics and extrinsics, and recognizing ink colors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If large robots are used to extend across printed surfaces, then the coverage area increases, but the dynamic loads on the printhead assembly increase causing positioning errors
Solution Approach 1:
The system uses cameras to capture images of the printed surface and calibration artifacts, then processes these images to detect positioning errors and nozzle outages. The detected information is fed back to the control system to compensate for positioning errors and rectify printing defects in real-time, maintaining precision despite large robot scale.
Solution Approach 2:
The patent replaces mechanical positioning measurement systems with optical measurement systems (cameras). Instead of relying solely on mechanical encoders and position sensors, the system uses image processing to detect the actual position of the printhead and calibration artifacts, converting mechanical positioning problems into optical measurement problems for higher precision.
2Measurement precision
If cameras are used for real-time defect detection, then printing quality monitoring improves, but calibration complexity increases
Solution Approach 1:
The system performs camera calibration before actual printing operations using predefined calibration artifacts with known geometric features. This preliminary calibration establishes the relationship between camera coordinates and robot coordinates, creating transformation matrices that simplify subsequent defect detection operations without requiring complex real-time calibration.
Solution Approach 2:
The patent introduces calibration artifacts as intermediary objects between the camera and the printed surface. These artifacts contain known patterns (circles, lines, corners) that serve as reference markers, making the calibration process more straightforward by providing easily detectable geometric features for coordinate transformation.
3Measurement precision
If multiple calibration artifacts are printed at known positions, then calibration accuracy improves, but the preprocessing time increases
Solution Approach 1:
The calibration process is segmented into distinct phases: printing calibration artifacts, capturing images, detecting artifact positions, calculating transformation parameters, and applying calibration. This segmentation allows for optimized processing at each stage and enables parallel processing of multiple artifacts simultaneously, reducing overall calibration time while maintaining accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present application is directed for use with a printhead assembly that is mounted to a robot. One application provides a system for inkjet printing on large objects such as commercial aircraft surfaces. The application provides for calibrating a camera with relation to the motion of the robot. The calibration occurs is achieved using one or more calibration artifacts printed by the inkjet system.