FOF-02-2016 - Machinery and robot systems in dynamic shop floor environments using novel embedded cognitive functions
The developed machinery and robot systems should lead to a significant impact in the following areas:
- Automation of previously manual production in order to bring European production plants in cheap labour countries back to Europe
- Strengthen global position of European manufacturing industry through the introduction of the new technologies related to machinery and robots with enhanced capabilities
- Strengthen the innovation potential of European manufacturing industry through the creation of new products made possible with the new developed technologies
- Reduction of 20% of set-up and new product adaptation costs, increasing efficiency
- Significant improvement in the adaptability of manufacturing systems.
In order to ensure a high impact, both standardisation and certification activities have to be addressed in the proposal.
Proposals should include a business case and exploitation strategy, as outlined in the Introduction to the LEIT part of this Work Programme.
This topic complements other call topics in this area funded under FOF-12-2017 a.ii and LEIT-ICT Robotics topics
Current production shop floors are organised in a fixed combination of sequential automated and manual tasks. Each station, in which one or more tasks are performed, is designed for optimal productivity, and the whole linear sequence of operations is as well optimised for productivity. This paradigm is efficient when production is set to the maximum capacity and the same tasks are repeated in the same way in each cycle. However, this does not scale well to other situations. The complexity and cost of shop floor organisation increases dramatically when it comes to flexible production or logistics, as for example when mixing different product models, and the cost for introducing a new product reference is also very high. Moreover, this model lacks the capacity to react to unexpected technical problems that may arise.
Future shop floors have to endorse flexibility and define networks in which a tight collaboration between humans, machines and robots is key for performance e.g. maintenance operations and changes in product set-up. Therefore the shop floors must be supported by enhanced perception capabilities including the ability to reason over the perceived environment. By using novel embedded cognitive functions, machinery and robots should be able to collaborate as network agents in a realistic semi-structured environment, being able to adapt their behaviour in order to give a response to unforeseen changes or situations. Furthermore, the cognitive capabilities will allow the machinery and robots to evolve from being programmed for a dedicated task to the handling of a multitude of different tasks.
Research activities should address at least three of the following areas:
- Perception as an integrated cognitive capability, considering collaborative perception (counting not only with on-board sensors, but also with the sensing capabilities available in the whole shop floor), scene understanding, reasoning and acting (active perception).
- Perception as a way to create intelligent, dexterous "universal" devices for handling or manipulation of products or tools (e.g. handling of soft or shape changing objects, non-task dedicated devices)
- Mobility as a key factor for flexibility: machinery and robot systems should not only be able to autonomously navigate in realistic changing scenarios, but also develop the competences to switch from environment level navigation to the accurate positioning required to complete the operations.
- Methods and technologies to eliminate physical barriers such as safety guarding or enclosures have already been developed, but lack in inherent safety of the overall system. Cognitive capabilities in order to guarantee safety at all times, including when the system is down (e.g. maintenance, failure) should be researched so that it is possible to open the way to certification.
- Adaptation through context awareness and reasoning, aiming at making machinery and robots aware of their surroundings, so that they can perceive and obtain information on the non-programmed and non-expected situations, and adapt their behaviour in order to better handle them, while taking into account safety aspects..
- Life-long learning and knowledge sharing tools, reducing to the minimum the initial programming efforts, and reusing the acquired abilities and competences over the existing machines.
Robots and machines should not be considered as individual agents, but will have to be part of an overall interactive network which should be defined and possibly standardised.
Proof of concept in terms of at least one demonstrator should be delivered before the end of the project, excluding commercially usable prototypes, but convincingly demonstrating scalability towards industrial needs taking into account age and gender aspects, and making a clear case for the safety of the worker under all circumstances.
Activities are expected to focus on Technology Readiness Levels 5 to 7 and to be centred around TRL6.
This topic addresses cross-KET activities.
The Commission considers that proposals requesting a contribution from the EU between EUR 4 and 6 million would allow this specific challenge to be addressed appropriately. Nonetheless, this does not preclude submission and selection of proposals requesting other amounts.