Skip to main content

Smart Assembly Robot with Advanced FUNctionalities

Periodic Reporting for period 1 - SARAFun (Smart Assembly Robot with Advanced FUNctionalities)

Reporting period: 2015-03-01 to 2016-08-31

Over the last decades, robots have being extensively used by industrial manufacturers, such as the automotive industry. According to the current paradigm industrial robots perform their assignments in constrained environments, heavily depending on hard automation that requires time-consuming (re-)programming performed by experienced software engineers.
The assembly application although considered a promising robotic application, it has been proven challenging to automate due to precise grasping requirements, part variations, operations requiring high precision (e.g. snap fits) or special motions (e.g. twist insertions), and wear and tear of the assembly equipment. The above restrictions limit the applicability of robotic assembly to a small fraction of the potential cases. Thus, even expensive, massively produced products are still assembled manually under harsh conditions with mobile phones and tablets being illustrative examples. There is also a clear trend towards a shorter product lifetime with the lead time for setting up a production line/cell being drastically reduced.
During recent years there have been several attempts of designing robots that are inherently safe and thus can work together with humans in mixed assembly lines. However, even in these setups, when a traditional programming approach is employed still several months are needed for accomplishing a new assembly. Such long integration times make the payback calculation for many potential customers challenging, whereas a desired dynamic production setting is prohibited.
Hence additional research is needed towards developing programming methods that will enable a complete integration of a new assembly task to be made in a much shorter period, e.g. a single day. A “single day integration” robotic solution for assembly of small parts is expected to be a disruptive game changer in the field of industrial manufacturing and will cause a re-evaluation of assembly manufacturing in all regions of the world, including those presenting higher-wages.
The SARAFun project has been formed to enable a non-expert user to integrate a new bi-manual assembly task on a robot in less than a day. This will be accomplished by augmenting the robot with cutting edge sensory and cognitive abilities required to plan and execute an assembly task demonstrated by the instructor.

For more information, publications, etc. visit:
During the first eighteen months of the SARAFun project, the consortium has produced significant progress in correspondence to the project’s overall work plan. The work in this period of the project focused on the analysis of the user requirements and the definition of the use cases, as well as on the overall SARAFun system specification and design. Moreover, the development efforts have been advanced in all key-areas of the SARAFun project, e.g. HRI interface, Key-frame extractor, physical human robot integration (pHRI) control, online motion generation, automatic finger design, slippage detection, assembly sequence generator, and folding assembly controller, along with the corresponding technical developments of the SARAFun hardware infrastructure. Several horizontal activities for diffusing project objectives, concepts and achievements to key stakeholders, and the general public have also taken place.
An outline of the project developments for the reporting period is provided below:
- Establishment of the overall SARAFun conceptual architecture and knowledge base
- Development of first version of SARAFun User Interface
- Development of teaching by visual demonstration framework
- The teaching by demonstration framework consists of three main stages; (1) Object detection, (2) Combined hand-object tracker, and (3) Automatic key-frame extraction.
- Initial work on physical human-robot interaction mode
- Implementation of path generation system
- Development of a methodology for teaching assembly forces
- System for automatic grasp planning and manipulation
- Learning the assembly of two parts by human demonstration
- Teaching of assembly with advanced physical human-robot interaction
- Integrated planning framework to plan grasps and optimize the finger design for industrial grippers
- Development of strategies to improve and maintain grasp stability for industrial grippers
- Transfer knowledge about human sensorimotor performance during assembly to the robot
- Development of effective multi-modal control of assembly strategies under uncertainties
- Validation of SARAFun project results in real assembly scenarios

Two real-world assembly tasks have been selected as benchmark applications for the results developed in the project. An overall system architecture has been developed to enable the teaching and executing of robot assembly tasks. 12 main functional components of the SARAFun system are defined and a dynamic analysis of the system is presented based on SysML activity and sequence diagrams.
Good progress has been made on the development of robot assembly skills targeting the assembly cases.
SARAFun Project Vision Figure