Ultrasonic Probe Path Learning for Accurate Robot-Guided Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurately positioning an ultrasonic probe for flaw detection is challenging when the exact position of the inspection target is unrecognized.
Innovation Solution
A control device for a flaw detection system that includes a probe and a robot arm, performing tracing, learning, and inspection processing to align the probe with the inspection target's shape, using robot path data and shape learning data to ensure precise contact and alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the probe is manually positioned for inspection, then the inspection can be performed, but the positioning accuracy is insufficient when the exact position of the inspection target cannot be recognized
Solution Approach 1:
The system performs tracing processing before inspection to pre-learn the shape and position of the inspection target. The robot arm moves the probe along the target surface to acquire shape data, which is then used to generate accurate robot path data for subsequent inspection operations, eliminating the need for manual positioning.
Solution Approach 2:
The system uses the learned shape data from tracing processing as feedback to automatically adjust and generate robot path data. This feedback mechanism allows the system to adapt to the actual target geometry and achieve precise probe positioning without manual intervention.
2Area of stationary object
If the robot arm is used to displace the probe, then the inspection coverage can be increased, but the alignment accuracy with the target surface becomes difficult to maintain
Solution Approach 1:
The system dynamically adjusts the robot path based on the learned shape data. The robot arm displaces the probe along a generated path that automatically adapts to the target's surface geometry, maintaining optimal contact and alignment throughout the inspection area while maximizing coverage.
Solution Approach 2:
The robot path data is generated by transforming the learned shape data into motion parameters. The system changes the robot's position and orientation parameters dynamically along the inspection path to maintain precise alignment with the target surface while covering the entire inspection area.
3Reliability
If the probe contact force is increased to ensure contact, then the reliability of contact is improved, but the risk of damaging the inspection target or probe increases
Solution Approach 1:
The system performs tracing processing to pre-learn the target shape and determine the optimal contact path before actual inspection. This preliminary action allows the system to identify gentle contact zones and plan a path that maintains reliable contact with minimal force, reducing damage risk.
Solution Approach 2:
The generated robot path data includes optimized contact force parameters. The system adjusts the probe's contact force dynamically along the path, applying sufficient force for reliable contact in rigid areas while reducing force in sensitive regions, thereby maintaining contact reliability without causing damage.
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
Enables accurate flaw detection by ensuring the probe is properly positioned and aligned with the target's actual shape, accounting for tolerances, thereby enhancing inspection accuracy.
Implementation Method 1
The inspection processing is processing for inspecting the inspection target by using a reflected wave of a transmitted ultrasonic wave
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A control device (40) of a flaw detection device uses, as a control target, a flaw detection device provided with a probe (28a) and a robot arm (26) that presses the probe against an inspection object. The control device executes tracing processing, learning processing, and inspection processing. In the tracing processing, the probe is moved along the inspection object while being brought into contact with the inspection object by operating the robot arm. In the learning processing, the shape of the inspection object is learned according to the position of the probe for every movement thereof in the tracing processing. In the inspection processing, the inspection object is inspected by using the reflected wave of the ultrasonic wave transmitted while displacing the probe brought into contact with the inspection object according to the shape of the inspection object learned by the learning processing.