COLEARNING-DEVELOPMENT OF TASK LEARNING TECHNIQUES FOR COBOTS BASED ON HUMAN INTERACTION AND REINFORCEMENT LEARNING ALGORITHMS (COLEARNING)
Generally speaking, robot programming has been greatly simplified in recent years. Manufacturers have developed graphic tools and gestural methods that facilitate the generation of programs in any type of robot. Obviously, each program is valid in a specific environment and for a specific task, so reprogramming is required if either of the two circumstances changes.
Therefore, one of the key challenges of deploying COBOTS (collaborative robots) in environments where human presence is important, such as manufacturing industries, is adapting to tasks that change frequently and unpredictably. In these circumstances, the robot should be able to quickly update the way it performs its task. In other words, they should be able to learn incrementally and adapt to the circumstances of both the environment and the task.
Beyond the programming systems that each robot manufacturer uses, which tend to simplify and facilitate programming, techniques can be implemented that accelerate robot learning to perform new tasks. And something that would be more important in certain types of work: they would allow facing tasks that are not exactly the same as the one learned, being able to assume variations that are part of the changing nature of the processed products, adapting the work cycle to the characteristics of each product individual.
The objective of the project is to develop a learning system for COBOTS that allows the reconfiguration of the program autonomously, depending on the variations of the process or product that cannot be fully considered in a standard programming routine. As a use case, it would be applied to a process with a lot of manual workload: the sanding / polishing of pieces of board and / or metal sheet, whose geometry differs slightly from one piece to another due to the forming process or due to the flexibility of the piece itself. material.
In this project, two different learning techniques will be used and compared to achieve the objective of self-reconfiguration of the work cycle: on the one hand, human supervision, through which COBOT can acquire new capabilities by demonstrating tasks; on the other hand, reinforcement learning through neural networks and the use of reward functions, which would allow to improve the performance of the task from the results that are obtained.
The project follows the lines set by the Smart Specialization Strategy for Research and Innovation in the Valencian Community (RIS3-CV), and the ERDF operational program of the Valencian Community for the period 2014-2020 in the fields of action contemplated in the Priority Axis 1, "Promote research, technological development and innovation". It is also part of Priority Axis 2 "Innovative Product" and Area of Specialization in field B "Advanced Manufacturing" and C "ICT", while the project will allow a substantial improvement in the automation of processes with a strong load of handwork.
Additionally, within Priority Axis 2 we have Sectorial Environment of Strategic Scope 2.1. “Consumer goods” that identifies important future needs such as “Developing personalized products that incorporate added value based on design and differentiated features, responding to individual customer needs” and the Specific Objective BC2 (in its first proposal) “ To promote new processes for the mass manufacturing of small series on demand, as well as to provide products with intelligent multifunctions and ICT solutions that allow their connection to the network, favoring the transformation of goods into services ”.
Non-economic R&D activity
The tasks included in this project are classified as “non-economic activity” based on the third section of the Community Framework on State Aid for R & D & I, 2006 / C323 / 01.
In accordance with the provisions of the aforementioned Community Framework, AIDIMME clearly distinguishes between economic activity and non-economic activity, and between their respective costs and financing. In addition, the project activities, as indicated in the Community Framework:
• They involve learning and improvement in the qualification of the research team working on the project;
• Improves knowledge and understanding of the research line where the project is included;
• The project has associated dissemination actions which are open and accessible to the business environment;
• The technology transfer derived from the results of the project is not of an economic nature, and if it were to be so (patents, licensing, etc.) any income generated would be used again in non-economic R&D activities of the center.
Project Number: 22000052
Grant Agreement: IMDEEA / 2020/22
Duration: From 01/04/2020 to 30/06/2021
Coordinated in AIDIMME by: SÁNCHEZ ASINS, JOSÉ LUIS
R&D Line: INDUSTRY 4.0
Objective Sectors by CNAE
CNAE: 3109 - Manufacture of other furniture - No. of target companies: 5517 (1305 in the Valencian Community)
CNAE: 2561 - Treatment and coating of metals - Number of target companies: 90 (56 from the Valencian Community)
CNAE: 3101 - Manufacture of office furniture and e - Number of target companies: 503 (106 in the Valencian Community)
CNAE: 3103 - Manufacture of mattresses - No. of target companies: 138 (38 in the Valencian Community)
CNAE: 3102 - Manufacture of kitchen furniture - Number of target companies: 652 (153 from the Valencian Community)
CNAE: 3299 - Other manufacturing industries nec - No. of target companies: 97 (46 from the Valencian Community)
Objective Sectors by Activity
FURNITURE - Number of target companies: 225 (93 of the Valencian Community)
Thanks to its support and signature of the "declaration of participation" the project has been funded.
DR. FRANZ SCHNEIDER, SA
MUEBLES ROMERO, SA
PUNT MOBLES XXI, SL
HURTADO RIVAS, SL
COMPANIES DIFFUSION R&D PROJECTS
They want to know first-hand the evolution of the project, and its progress to the final result.
COMPANIES TRANSFER KNOWLEDGE
They will implement technologies, develop strategies or look for new models based on the results.