Trajectory Creation Towards Fast Skill Deployment in Plug-and- Produce Assembly Systems: A Gaussian-Mixture Model Approach
Title: Trajectory Creation Towards Fast Skill Deployment in Plug-and- Produce Assembly Systems: A Gaussian-Mixture Model Approach
Authors: Melanie Zimmer (Intelligent Automation Centre, Loughborough University); Ali Al-Yacoub (Intelligent Automation Centre, Loughborough University); Pedro Ferreira (Intelligent Automation Centre, Loughborough University); Niels Lohse (Intelligent Automation Centre, Loughborough University);
Year: 2019
Citation: Zimmer, M., Al-Yacoub, A., Ferreira, P., Lohse, N., (2019). Trajectory Creation Towards Fast Skill Deployment in Plug-and- Produce Assembly Systems: A Gaussian-Mixture Model Approach. UK-RAS19 Conference: “Embedded Intelligence: Enabling & Supporting RAS Technologies” Proceedings, 87-90. doi: 10.31256/UKRAS19.23
Abstract:
In this paper, a technique that reduces the changeover time in industrial workstations is presented. A Learning from Demonstration-based algorithm is used to acquire a new skill through a series of real-world human demonstrations in which the human shows the desired task. Initially, the collected data are filtered and aligned applying Fast Dynamic Time Warping (FastDTW). Then the aligned trajectories are modelled with a Gaussian Mixture Model (GMM), which is used as an input to generate a generalisation of the motion through a Gaussian Mixture Regression (GMR). The proposed approach is set into the context of the openMOS framework to efficiently add new skills that can be performed on different workstations. The main benefit of this work in progress is providing an intuitive, simple technique to add new robotics skills to an industrial platform which accelerates the changeover phase in manufacturing scenarios.