AI Fastener Identification for Adaptive Aircraft Removal Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in industrial robotics is the variability in the arrangement of aircraft fasteners, which complicates the design of automated systems for their removal, as each aircraft part has a unique set of fasteners with different orientations, making it difficult to use one set as a template for another, and requires precise path and trajectory planning for proper removal.
Innovation Solution
A system that uses dynamic path planning and trajectory planning, guided by a user interface (UI) to receive point cloud data, segment objects, identify targets like holes or fastener heads, estimate their 3D position, and generate paths for a robot's end effector, incorporating artificial intelligence (AI) and machine learning (ML) for precise robotic operations, including fastener removal, and allows for updating of multi-dimensional representations for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional template-based approaches are used for fastener removal, then the system design is simplified, but the system cannot adapt to the variability in fastener arrangements across different aircraft parts
Solution Approach 1:
The system transitions from static template-based approaches to dynamic adaptive planning. The path and trajectory planning are generated in real-time based on detected fastener positions and orientations, allowing the robotic system to adapt to each unique aircraft part configuration rather than relying on pre-defined templates.
Solution Approach 2:
The system performs self-adaptation by automatically detecting fastener arrangements and generating appropriate removal paths without requiring external reprogramming or manual intervention for each part variation. The robotic system serves itself by learning and adapting to different configurations autonomously.
2Manufacturing precision
If dynamic path and trajectory planning is implemented for each fastener, then precise removal is achieved, but the computational complexity and processing time increase
Solution Approach 1:
The system performs preliminary detection and classification of fasteners before generating removal paths. By identifying all fasteners, their types, positions, and orientations in advance, the system prepares the necessary information for efficient path planning, reducing the computational burden during the actual removal operation.
Solution Approach 2:
The complex task of fastener removal is segmented into distinct phases: detection, classification, path planning, and execution. Each phase handles specific sub-tasks independently, allowing for optimized processing in each segment and preventing the need to recalculate everything from scratch.
3Adaptability or versatility
If AI and ML models are integrated for fastener identification, then the system can handle complex geometries and variations, but the computational resources and processing requirements increase
Solution Approach 1:
The system replaces traditional mechanical vision systems and manual programming with AI and ML-based perception. These intelligent models automatically learn to identify fasteners and their characteristics from images, substituting complex mechanical processing with more efficient computational patterns that require fewer resources for similar or better performance.
Data Source
AI summary
Aspects of the disclosure are directed towards artificial intelligence-based modeling of target objects, such as aircraft parts. In an example, a system initially trains a machine learning (ML) model based on synthetic images generated based on multi-dimensional representation of target objects. The same system or a different system subsequently further trains the ML model based on actual images generated by cameras positioned by robots relative to target objects. The ML model can be used to process an image generated by a camera positioned by a robot relative to a target object based on a multi-dimensional representation of the target object. The output of the ML model can indicate, for a detected target, position data, a target type, and/or a visual inspection property. This output can then be used to update the multi-dimensional representation, which is then used to perform robotics operations on the target object.