Reactive systems are ubiquitous computer
systems which maintain an ongoing interaction with an external environment,
responding with appropriate actions to the external stimuli, and which
often perform critical activities such as controlling industrial plants,
managing financial transactions, scheduling time-critical operations,
running e-commerce and web applications. With the increase of the dependency
of our society from computer applications and reactive systems their correct
behavior is becoming more and more critical, however their verification
is particularly difficult. Practical examples of reactive systems in
fact often consists of several sub-components that interact with an environment.
Furthermore, their specification may require heterogeneous data and may
depend on parameters whose values cannot be fixed a priori. Also, in
many cases the set of possible states is infinite. Any of these aspects
represents a big obstacle for the applicability of the state-of-the-art
verification technology.
The scientific global goal
of the project is to use the notion of `constraint', supported by
abstract interpretation and type theory, as the unifying concept for
the specification and automated verification of complex reactive systems.
|