Synthesis of Safety Specifications for Probabilistic Systems
Ohlmann, Gaspard, Court, Edwin Hamel-De le, Belardinelli, Francesco
–arXiv.org Artificial Intelligence
Ensuring that agents satisfy safety specifications can be crucial in safety-critical environments. While methods exist for controller synthesis with safe temporal specifications, most existing methods restrict safe temporal specifications to probabilistic-avoidance constraints. Formal methods typically offer more expressive ways to express safety in probabilistic systems, such as Probabilistic Computation Tree Logic (PCTL) formulas. Thus, in this paper, we develop a new approach that supports more general temporal properties expressed in PCTL. Our contribution is twofold. First, we develop a theoretical framework for the Synthesis of safe-PCTL specifications. We show how the reducing global specification satisfaction to local constraints, and define CPCTL, a fragment of safe-PCTL. We demonstrate how the expressiveness of CPCTL makes it a relevant fragment for the Synthesis Problem. Second, we leverage these results and propose a new Value Iteration-based algorithm to solve the synthesis problem for these more general temporal properties, and we prove the soundness and completeness of our method.
arXiv.org Artificial Intelligence
Nov-21-2025
- Country:
- Asia > India
- Kerala > Thiruvananthapuram (0.04)
- Europe
- France > Occitanie
- Haute-Garonne > Toulouse (0.04)
- Iceland > Capital Region
- Reykjavik (0.04)
- Slovenia > Drava
- Municipality of Benedikt > Benedikt (0.04)
- United Kingdom > England
- Greater London > London (0.04)
- France > Occitanie
- North America > United States (0.04)
- South America > Argentina
- Pampas > Buenos Aires F.D. > Buenos Aires (0.04)
- Asia > India
- Genre:
- Research Report (0.63)
- Technology: