Dynamic reconfigurability of collaborative robotic systems

Project content

In contrast to conventional industrial robots, collaborative robots do not require protective fences for safety reasons since they react to occurring forces. In this way, human-robot-collaboration (HRC) can be realized in principle and ensure the necessary personal safety, provided that potential contact situations are determined, measured and compared with the limit values of ISO/TS 15066. Due to the current significant limitation of the newly required risk assessment and measurement-based assessment of each application and change, the essential flexibility of a classical industry robot which makes the system adaptable anytime is rendered void in a collaborative robot. In addition, collaborative robots (stationary and mobile) should preferably be used in dynamic environments, which means that a robot must adapt dynamically as a result of changing requirements. The project addresses the current needs of the business community in HRC applications and tries to guarantee the necessary personal safety with a far-reaching increase in system flexibility. In contrast to many other experiments, it is not necessary to use imaging methods but instead mainly industrially tested protective devices and safety functions are to be considered.

With DR.KORS we aim to define modification limits for a basic application on a well-known robot system. Thus, changes to the system and to the application, in various dimensions and limits known to the user, can be carried out within a certain framework. The safety assessment and approval of a desired modification is carried out by a software framework which evaluates the guaranteed compliance with the personal safety on the basis of the underlying safety system models. These models are filled with parameters during commissioning of the plant, which are determined, among other things, by means of biomechanical measurement systems or which are robot-dependent. Statistical models also provide information on the personal safety of protective equipment. In addition to these extrinsic variables, work system parameters are also generated, which provide real-world boundaries for intrinsic modelling. The comprehensive, automated interpretation of personal safety makes it possible to achieve a dynamic, multidimensional system and application modification of collaborative robotic systems.

Project information
  • Financing agency: FFG (Austrian Research Promotion Agency)
  • Call: FFG programme "Production of the Future"
  • Project period: 07/2018 - 12/2020