Parallel Parameter Synthesis for Multi-affine Hybrid Systems from Hybrid CTL Specifications.

Logo poskytovatele

Varování

Publikace nespadá pod Filozofickou fakultu, ale pod Fakultu informatiky. Oficiální stránka publikace je na webu muni.cz.
Autoři

ŠMIJÁKOVÁ Eva PASTVA Samuel ŠAFRÁNEK David BRIM Luboš

Rok publikování 2020
Druh Článek ve sborníku
Konference Computational Methods in Systems Biology. CMSB 2020. Lecture Notes in Computer Science, vol 12314
Fakulta / Pracoviště MU

Fakulta informatiky

Citace
www https://link.springer.com/chapter/10.1007/978-3-030-60327-4_15
Doi http://dx.doi.org/10.1007/978-3-030-60327-4_15
Klíčová slova Hybrid systems; Parameter synthesis; Rectangular abstraction; Semi-symbolic; Hybrid CTL
Popis We consider the parameter synthesis problem for multi-affine hybrid systems and properties specified using a hybrid extension of CTL (HCTL). The goal is to determine the sets of parameter valuations for which the given hybrid system satisfies the desired HCTL property. As our main contribution, we propose a shared-memory parallel algorithm which efficiently computes such parameter valuation sets. We combine a rectangular discretisation of the continuous dynamics with the discrete transitions of the hybrid system to obtain a single over-approximating semi-symbolic transition system. Such system can be then analysed using a fixed-point parameter synthesis algorithm to obtain all satisfying parametrisations. We evaluate the scalability of the method and demonstrate its applicability in a biological case study.
Související projekty:

Používáte starou verzi internetového prohlížeče. Doporučujeme aktualizovat Váš prohlížeč na nejnovější verzi.