This paper addresses the exact computation of the set of reachable states of a strongly linear hybrid system. It proposes an approach that is an extension of classical state-space exploration. This approach uses a new operation, based on a cycle analysis in the control graph of the system, for generating sets of reachable states, as well as a powerful representation system for sets of values. The method broadens the range of hybrid systems for which a finite and exact representation of the set of reachable states can be computed. In particular, the state-space exploration may be performed even if the set of variable values reachable at a given control location cannot be expressed as a finite union of convex regions. The technique is illustrated on a very simple example.
CITATION STYLE
Boigelot, B., Bronne, L., & Rassart, S. (1997). An improved reachability analysis method for strongly linear hybrid systems (Extended abstract). In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1254, pp. 167–178). Springer Verlag. https://doi.org/10.1007/3-540-63166-6_18
Mendeley helps you to discover research relevant for your work.