The overarching RSRG research objectives are ambitious and long-term:
Our group is sometimes known as the RESOLVE group, because one of the results of our research has been the development of the framework, research language, and component design discipline called RESOLVE (a.k.a. Resolve).
The key technical issue that drives our work is the requirement for modular verification of software behavior. By this, we mean that it must be possible to prove once-and-for-all that an implementation component satisfies an abstract specification of behavioral properties -- and then to put the implementation code away and henceforth to rely only on that specification when reasoning about related properties of the component's clients. Our results demonstrate that this requirement impacts not only software component design, but virtually every facet of software engineering and programming languages and even the foundations of computation.