Integrating Cyber-Physical Systems in Non-rigid Assemblies: A Composites Manufacturing Case Study

Despite the advances in robot control and Internet of Things (IoT) services, manufacturing operations involving the handling of non-rigid products or workpieces largely remain manual. The dynamic distortion of flexible objects underlines limitations in cognitive robot dexterity as well as the necessity for digital twin platforms where modules for (co-) manipulation planning are applied. These planning modules cannot demonstrate their full potential in terms of performance and resilience without an accurate reconstruction of the physical workstation. This need becomes more apparent in unstructured environments where human intervention and improvisation may occur, or when the assembled product exhibits stochastic behavior. This paper discusses a digital-twin framework that can address the challenges associated with non-rigid assemblies and fenceless human-robot co-existence. A use case coming from the automotive industry is used for validating the framework, by enhancing aspects of safety, interaction, perception, cognition, orchestration and performance of a hybrid cell for composites manufacturing.