Automated Aircraft Assembly System with Force Feedback Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The assembly of large vehicle components, such as aircraft fuselage sections, often requires human intervention due to limitations in automated positioning systems, which can lead to inefficiencies and increased residual stresses.
Innovation Solution
A cyber-physical production system that integrates positioner units, force sensors, and vehicle components with data storage and communication capabilities, controlled by a multi-agent system using machine learning and simulation models to optimize positioning and minimize deviations from nominal positions while managing reaction forces and stresses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated positioning systems are used for assembly, then productivity is improved, but the system requires human intervention when components exceed predefined boundaries, reducing automation extent
Solution Approach 1:
The system enables self-service automation by allowing the automated positioning system to autonomously handle components that exceed predefined boundaries. The machine learning model automatically adjusts positioning parameters and trajectories without human intervention, enabling the system to service itself and maintain continuous automated operation.
Solution Approach 2:
The system implements continuous feedback loops where sensors monitor component positions and reaction forces in real-time. This feedback is fed into the machine learning model, which dynamically adjusts positioning parameters to keep components within acceptable boundaries, enabling fully automated decision-making and eliminating the need for human operators to intervene.
2Adaptability or versatility
If iterative positioning processes with human operators are used, then adaptability is improved, but loss of time increases due to manual intervention
Solution Approach 1:
The machine learning model performs self-learning and self-adjustment during the assembly process. It automatically adapts to different component variations and boundary conditions by processing sensor data and adjusting positioning parameters in real-time, eliminating the need for human operators to manually intervene and maintain assembly flow continuity.
Solution Approach 2:
The system performs preliminary positioning adjustments using the machine learning model before components are mounted. The model predicts optimal positioning trajectories and makes preemptive adjustments to prevent boundary violations, thereby avoiding iterative corrections and reducing overall assembly cycle time while maintaining adaptability.
3Productivity
If automated positioning systems operate without human intervention, then productivity is improved, but manufacturing precision may deteriorate due to system boundaries
Solution Approach 1:
The machine learning model dynamically changes positioning parameters such as trajectories, speeds, and forces based on real-time sensor feedback and component characteristics. This allows the automated system to adapt positioning parameters to maintain high precision while operating continuously without human intervention, effectively resolving the contradiction between productivity and precision.
Solution Approach 2:
The system implements multi-level feedback mechanisms including force sensors that monitor reaction forces at mounting points and position sensors that track component locations. This feedback is continuously processed by the machine learning model to make real-time adjustments to positioning parameters, ensuring manufacturing precision is maintained even as productivity increases through continuous automated operation.
4Measurement precision
If more sensors and data storage are integrated into vehicle components, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system employs universal sensor platforms and standardized data interfaces that can be applied across different vehicle component types. The machine learning model serves multiple functions including positioning control, stress analysis, and quality assessment, thereby improving measurement precision without proportionally increasing device complexity through functional integration.
Solution Approach 2:
The machine learning model acts as an intermediary that processes and integrates data from multiple sensors (position sensors, force sensors, status sensors). This intermediary layer simplifies the overall system architecture by consolidating data processing functions and providing a unified interface between sensors and the positioning control system, thereby improving measurement precision while managing device complexity.
Data Source
Figure 1a~2
Figure 3
Figure 4a~4b
AI summary
The present invention pertains to a production system (50) for the automated assembly of vehicle components (1), in particular for the automated assembly of structural components of an aircraft or spacecraft (100). The production system (50) comprises vehicle components (1) being provided with status sensors (10), each status sensor (10) being configured to determine status data (12) of the respective vehicle component (1), positioner units (2) being configured to grip the respectively associated vehicle component (1) at mounting points (7) and move the respectively associated vehicle component (1) into an assembly position (3), a position-measurement system (4) being configured to determine the assembly position (3) of each vehicle component (1), force sensors (5) being configured to determine at least one of reaction forces and moments of each gripped vehicle component (1) at the mounting points (7) in the assembly position (3), and a computer-based control system (30) being in data communication with the vehicle components (1), the positioner units (2), the position-measurement system (4) and the force sensors (5), and being configured to control the positioner units (2) based on the determined status data (12), the determined assembly positions (3) and the determined reaction forces and moments of the vehicle components (1).