Maximal Safe Controllable Domain
Abstract
The semantic indefinitely safe domain is the greatest controlled-safe fixed point.
Theorem 1.1 (The indefinitely safe domain is the greatest fixed point).
Proof. Machine-checked in Lean as D5/S3/ConceptDynamics/Control/MaximalSafeControllableDomain.maximal_safe_controllable_domain (✓ std3). ∎
Source. Repository-derived.
Commentary.
A control system supplies a state-dependent action type and a nonempty successor set for every available action. The controlled predecessor uses existential action choice and universal successor containment.
The indefinitely safe set is constructed semantically: a state must lie in some subset of the safe states that offers a confining action at each of its states. It is not defined as a fixed point.
Independently, the displayed monotone operator intersects the safe set with the canonical controlled predecessor. Knaster-Tarski identifies its greatest fixed point with the semantic indefinitely safe set.
The remaining public clauses expose the confining action, indefinite-safety inclusion, and converse maximality. Repository searches found no exact theorem; Mathlib’s greatest-fixed-point laws are applied directly.
References
- Truth anchor:
D5/S3/ConceptDynamics/Control/MaximalSafeControllableDomain.maximal_safe_controllable_domain - Dependency: D5/S3/ConceptDynamics/Control/FiniteHorizonReachability