Search for: All records

Cyber-physical system (CPS) testbeds are critical research tools for advancing safety-critical technologies, from autonomous vehicles to smart grids. As CPS grow more complex, adaptive, and interconnected, testbeds must evolve in kind—yet the architectural assumptions guiding their design remain fragmented across domains. This paper introduces a morphological framework that enables comparison and synthesis across traditionally siloed CPS fields. Drawing on a structured review and a feature abstraction process, we derive 25 architectural dimensions and use them to classify 113 CPS testbeds spanning a wide range of use cases. Cluster analysis reveals three dominant design archetypes characterized by centralized control, reconfigurability, and virtualized operation, and highlights convergence in user interaction mechanisms and the decision-making architectures that govern testbed behavior. We quantify configurational rigidity, expose underexplored regions of the design space, and identify architectural opportunities in emerging CPS research domains. By reconceiving testbeds as design artifacts, this work opens new avenues for architectural innovation in CPS testbed research.