Analysis of the combination of machine vision and artificial intelligence technologies for on-line inspection
Program: AEIs. 2020
Referencia: AEI-010300-2020-126
Resolution Date: 03/11/2020
Execution date: 07/01/2020 – 03/31/2020-31/03/2021
The objective of the project is the development of a feasibility study for the development of an in-line quality control system based on expert systems (machining learning, deep learning, etc.) and artificial vision techniques mounted on a robot tip, which can be directly integrated into the production line process, taking into account that it is a changing environment in terms of lighting conditions and part geometry (evolution of the part in its life cycle).
Specifically, the development of the feasibility study will be carried out using as a practical example one of DANA’s production lines and taking one of the most manufactured references (it will allow to have a greater number of images for the training of the system). A laboratory environment will be designed to replicate as far as possible the system conditions (realistic sensor positions taking into account the geometry of the system) necessary to analyze the feasibility of the solution and to have a tool that allows a priori to establish which of the multiple types of expert systems could be better adapted to the solution of the problem.
In order to achieve this objective, the following specific objectives are proposed:
Objective 1. Perform a complete use case definition that can be used to determine the needs of the solution to be used. The definition of the use case and specifications will be carried out using ITAINNOVA’s own methodology called MDR (Robust Design Methodology), which allows the capture of information in a systematic and consistent way so that the collection of specifications is defined and delimited.
Objective 2. Laboratory design of the use case. After defining the use case and specifications, a prototype of the inspection station will be made in the laboratory as realistic as possible using a robotic cell in the laboratory, which will ensure that subsequent solutions tested on this cell can be used in the final solution.
Objective 3. Experiment with different learning-based expert system solutions to analyze which of them is best suited to the specific use case problem. For this purpose, in addition to the different techniques used for the development of these systems, other parameters such as the image acquisition elements and their conditioning will be taken into account.
Objective 4. Verify the correct operation of the selected solution/s in a real environment, for this purpose, the possibility of performing a test on the use case line itself will be analyzed and/or images taken in the real environment will be used.
Objective 5. Disseminate the results of the project among companies in the sector to promote the use and application of technologies within the framework of Industry 4.0.
ARAGON AUTOMOTIVE CLUSTER ASSOCIATION
TECHNOLOGICAL INSTITUTE OF ARAGON
TOTAL PROJECT BUDGET: 73,894 €.
TOTAL BUDGET ITAINNOVA: 52,938 €.
TOTAL PROJECT GRANT: 50,000 €.
TOTAL FINANCING ITANNOVA: 35,330 €.
The VIAT project has been supported by the Ministry of Industry, Trade and Tourism (Agrupaciones AEIs Program).
