The French ANR Project COROUSSO:

Robot Control for Machining and Welding


Project type:

Funded by the ANR (French Research Funding Agency)

Funding:

1,106,942 Euros

Project ID:

ANR-10-SEGI-0003

Schedule:

Oct. 15, 2010 - Oct. 14, 2014

Partners:

LCFC, IRCCyN, Institut de Soudure, Europe Technologies

Also supported by:

the EMC2 and Materalia French clusters for competitiveness.

Takes part in the EquipEx RobotEx.

Project webpage:

http://www.corousso.sitew.com/#Objectifs.A

Objectives of the COROUSSO project

COROUSSO project deals with the modelling and control of robots for machining operations of large composite parts and friction stir welding (FSW). As a matter of fact, those machining and welding operations are usually realized by means of expensive custom-made machines. As a consequence, manufacturers (mainly aeronautical manufacturers) require new devices to allow them to decrease the manufacturing cost as well as to improve the machining quality. It appears that the robotization of the manufacturing processes is a relevant solution.

However, industrial robots are usually not stiff enough for machining operations and FSW and the quality of the resulting machined or welded parts does not respect the specifications. Accordingly, the project aims to propose three strategies to improve robot machining quality. First, the flexibilities of the robots will be taken into account in their control loop. Then, the uncertainties will be also considered in the control loop of the robots in order to obtain a robust control. Finally, the influence of the path (part to be welded or machined) placement on the machining quality will be studied.

This project will also focus on the elastodynamic performance of the robots. Indeed, the wrench exerted on the end-effector of the robot due to manufacturing or the welding process can generate some distortions of the robot architecture as well as some vibrations. In order to decrease those distortions and vibrations, the elastodynamic model of the robot will be implemented in its control loop. The parameters of the elastodynamic model will be also identified experimentally beforehand. Moreover, the distortions and vibrations of the robot should be also reduced by means of force feedback control and a force sensor that will be installed between the robot end-effector and the tool.

It is apparent that the better the modelling of the robot and its environment, namely the production process, the better the performance of the robot. Therefore, the kinematic and dynamic modelling of the FSW process and the manufacturing process used to machine large composite parts will be determined in the scope of this project. The partners of the project already have good knowledge of those processes but some research work has to be done in order to implement the obtained models in the control loop of the robot. This will be a major contribution of COROUSSO project.

The knowledge of COROUSSO partners fit well with the project. The IRCCyN (Nantes) is well known in the field of robotics and manufacturing and mainly for its research works on robot design, modelling and control. The LCFC (Metz) has contributed a lot the last few decades in the fields of modelling and optimization of production processes like FSW. Besides, LCFC researchers took part in some industrial projects on robot control. The two industrial partners, i.e., ``l’Institut de Soudure’’ and ``Europe Technologies’’ are well known for your research works on production process and development of new manufacturing techniques.

Finally, the COROUSSO project aims to provide theoretical and experimental results on the development of robotic cells dedicated to machining of large composite parts FSW. The technology transfer in the production domain will be realized by means of experimental tests conducted by the industrial partners to highlight critical operations. The contributions of the project should also help us robotize other production process later.