Nonlinear dynamical systems & control theory

3.5. The exponential contraction flow in 1-d🔗

We consider the flow of the vector field f : ℝ → ℝ given by f(x) = r x for some fixed r : ℝ.

🔗def
smulFlow (r : ) : Flow
smulFlow (r : ) : Flow

The flow of the vector field x r x.

Even without knowing the explicit solution of the ODE, it is easy to see that the function V(x) = x ^ 2 is Lyapunov function if r ≤ 0.

🔗theorem
isLyapunov_sq_smulFlow {r : } (hr : 0 r) : IsLyapunov (fun x x ^ 2) (smulFlow (-r)).toFun
isLyapunov_sq_smulFlow {r : } (hr : 0 r) : IsLyapunov (fun x x ^ 2) (smulFlow (-r)).toFun

The function x x ^ 2 is a Lyapunov function for the system d/dt x = (-r) x.

From this it follows that the origin is globally asymptotic stable:

🔗theorem
isStableOn_smulFlow {r : } (hr : 0 r) : (𝓝 0).IsStableOn (smulFlow (-r)).toFun (Set.Ici 0)
isStableOn_smulFlow {r : } (hr : 0 r) : (𝓝 0).IsStableOn (smulFlow (-r)).toFun (Set.Ici 0)

The origin is stable under the forward flow of d/dt x = r x

🔗theorem
tendsto_smulFlow {r : } (hr : 0 < r) (x : ) : Filter.Tendsto (fun x_1 (smulFlow (-r)).toFun x_1 x) Filter.atTop (𝓝 0)
tendsto_smulFlow {r : } (hr : 0 < r) (x : ) : Filter.Tendsto (fun x_1 (smulFlow (-r)).toFun x_1 x) Filter.atTop (𝓝 0)

The origin is globally asymptotic stable under the forward flow of d/dt x = r x