An extendable framework for intelligent and easily configurable skills-based industrial robot applications

Abstract

Modern, flexible, and easy-to-use robotic technologies have the potential to support companies to increase their productivity within today’s dynamic and volatile production. In this context, we introduce a skills-based software framework that makes it possible to configure the functional capabilities of industrial robots flexibly. In addition, we have structured the software framework into three consecutive expansion stages. In this way, it is possible to expand the robot’s reasoning capabilities step by step so that the robot is enabled to be instructed at higher abstraction levels and to process increasingly complex tasks. The contribution of our work is the further development of previous approaches and ideas from the research field of skills-based industrial robotic frameworks by considering new and previously unaddressed design issues within the structure of our software framework. We demonstrate the application of the framework using the example of an industrial robot for assembling a diverse range of LEGO products. The example of use consists of three consecutive scenarios. To begin with, the robot assembles different predefined product variants. Subsequently, we extend the robot application in a step-by-step manner to allow the robot to execute more and more complex tasks until it can finally plan individual tasks autonomously. On the one side, our approach shows how to enable companies with little robotic experience to start developing robotic applications and thereby gain further expertise. On the other side, by using this approach the effort and time for developing industrial robot applications will be reduced in the long term.