Simulation of hybrid systems from natural-language requirements

Cyber-physical systems are characterised by a massive and tight interaction between computer systems and physical components. Hybrid systems provide an abstraction for modelling cyber-physical systems by featuring the integration of discrete and continuous behavioural aspects. Simulation is an important tool for validating hybrid system models, which are often too complex to be treated using other validation and verification techniques. Motivated by the industrial need for such tools, we propose a strategy (h-NAT2TEST) for simulation of hybrid systems from natural-language requirements. Using the proposed approach, one writes the system specification using a controlled natural language, from which an informal semantics is automatically inferred based on the case grammar theory. Then, a formal representation is built considering a model of hybrid data-flow reactive systems (h-DFRS). Finally, in order to allow for rigorous simulation, an Acumen specification is derived from the h-DFRS model. Simulation is supported by the Acumen modelling environment. A DC-DC boost converter is used as a case study to illustrate the overall approach.