SDC2

The common goal of this team would be to meet the requirements of a dynamic process, in this case the process of disassembling a product in a recycling context. The distribution of functions between human and robot must indeed adapt to product variability and in particular to related uncertainties such as wear and tear. The robot must therefore act as a partner, able to meet the needs of the human through its complementary capabilities (integrative cooperation) or by sharing functions similar to those that the human can perform but not alone (augmentative cooperation). The implementation of this cooperation will not only ensure compliance with the normative rules in force, but also the comfort and interest that humans may find in it. This project is in line with regional objectives relating to the circular economy, national objectives relating to Industry 5.0 and human-centered design, and would be integrated into the SDC2 (Smart Disassembly Cell for Circularity) project supported by the CARNOT ARTS, CETIM and Energie du Futur Institutes.

A method for designing and evaluating a Human-Robot system dedicated to disassembling products of varying size and complexity will be proposed. Drawing on work from other application fields, the method will be adapted to the specific case of disassembly, to identify the know-how and know-how-cooperation of human operators and robots to jointly meet the needs associated with a specific product. In particular, the method will highlight the phases of training agents, preparing the activity (planning and allocation of functions), supervising and controlling the activity until the end of, and then capitalizing on knowledge about the product and the individual and cooperative skills and abilities of the agents.

Two use cases will be studied. The method will be deployed at LISPEN (Arts et Métiers Lille) with the aim of: 1) carry out a comparative analysis of the influence of different levels of exploitation of human know-how on performance indicators (in the training and operating phases) for a case of disassembly of pneumatic cylinders and/or electric motors, 2) open the methodology to the case study of large, complex products (large household appliances) without a priori knowledge of disassembly modalities. It will also be applied at LAMIH (UPHF) as part of a human-robot cooperation (Yumi) within a reconfigurable robotized cell for disassembling small electric motors.