Methods and tools for the verification of plug-and-produce robots in distributed manufacturing systems

The competitiveness of manufacturing companies in modern markets increasingly depends on cost-effective flexible automation technologies. Reconfigurable manufacturing systems can accommodate changes to introduce new product variants, adjust production capacity or recover from disruptions. Robots, inspection devices, material handling units and other pieces of equipment can be added to the production system, changed or removed. This level of flexibility requires extensive human intervention and, therefore, it is time-consuming and expensive.

Plug-and-produce systems introduced a significant degree of autonomy. New modules can be “plugged” into a distributed production system and be automatically integrated and configured. However, the verification of such systems is challenging due to their dynamic architecture and the high variability of possible hardware and software configurations. Furthermore, it is highly desirable to test accurately as much as possible offline, i.e. in a simulation environment, without operating the real robots.

Plug-and-produce systems enable collaboration even between heterogeneous components, which must trust each other in order to cooperate. Ensuring a reliable cooperation in these reconfigurable systems is the key to unlocking their full potential and enabling industries to develop business models confidently.

This project aims to develop novel methods and software tools for:

1. Automated generation of test cases for plug-and-produce robots in distributed manufacturing systems.
2. Offline simulation of plug-and-produce behaviour.
3. Analysis of the reliability of plug-and-produce robots in dynamic architectures.

It is anticipated that the automated generation of test cases will use model-based and search-based software testing techniques.

How to apply

Formal applications can be made via the University of Bradford web site.