(3.6), p. 15 — under C1 the errors (3.5) obey e_x(k+1) = (M(k) ⊗ I_p) e_x(k)
ProvedSubSuperStoch.AsyncTrack.error_systemConsider the asynchronous leader–follower network (3.1)–(3.3) with step size , gain and arbitrary communication instants. Assume condition C1: the signed digraph is structurally balanced with respect to the partition , . Define the error coordinates (3.5)
Then along every trajectory, for all and every follower ,
which is the error system written coordinate-wise.
The identity converts bipartite tracking (followers in approach , followers in approach ) into the convergence to zero of the products .
Formalization Note The stacked vector and the Kronecker product are replaced by the equivalent coordinate-wise identity on the blocks . The gain bound (3.7), , and the asynchronous-instant condition (2.10) are not assumed: the identity is algebraic and holds without them.
import Mathlib import Definitions.Def_SubSuperStoch_AsyncTrack_SignedDigraph import Definitions.Def_SubSuperStoch_AsyncTrack_AsyncModel
namespace SubSuperStoch.AsyncTrack
/-- (3.5)–(3.6), p. 15: under condition C1 (with partition `𝒱₁ = {v₀} ∪ V₁`), every trajectory of
(3.1)–(3.3) satisfies the error system `e_x(k+1) = (M(k) ⊗ I_p) e_x(k)`, written coordinate-wise:
`e_i(k+1) = ∑_j [M(k)]_{ij} e_j(k)`. -/
theorem error_system {n p : ℕ} (a : Matrix (Fin n) (Fin n) ℝ) (b : Fin n → ℝ)
(τ ψ : ℝ) (s : Fin n → ℕ → ℕ)
(V₁ : Set (Fin n)) (hC1 : StructBalanced a b V₁)
(x0 : ℕ → EuclideanSpace ℝ (Fin p)) (x : Fin n → ℕ → EuclideanSpace ℝ (Fin p))
(hrun : IsRun a b τ ψ s x0 x) :
∀ k i, errVec V₁ x0 x i (k + 1) = ∑ j, Mmat a b τ ψ s k i j • errVec V₁ x0 x j k := by sorry
end SubSuperStoch.AsyncTrack
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.