Manufacturers are looking for intelligent solutions to increase quality and productivity. Smart manufacturing envisions production empowered by autonomous robots that can complete tasks intelligently, with the focus on adaptability, flexibility, and versatility. In such systems, agile tasking plays an important role, as it is critical for robots to be quickly tasked to perform an operation. However, task agility is not limited to the speed of tasking robots, but also includes other features such as handling task failure, planning for new goals, interchangeability of data and task plans between different robots, and adapting to dynamic environments. Because robot task agility requires sophisticated dynamic and continuous planning and replanning, the Gwendolen agent programming language was chosen to evaluate as the agile robot planner. In this paper, we develop a manufacturing kitting case study and provide a list of kitting performance metrics to evaluate performance. The case study uses Gwendolen, Canonical Robot Command Language (CRCL), Robot Operating System (ROS) and Gazebo software components in combination to simulate and evaluate kitting. We explore the strengths of Gwendolen agile tasking to assess the operation against the kitting performance metrics.

This content is only available via PDF.
You do not currently have access to this content.