A team of researchers has shown that linear systems with continuously varying time delays can be stabilized over wider delay ranges than before. Their approach rewrites the delay differential equation as a feedback loop between a fixed‑delay dynamical core and a norm‑bounded uncertainty that captures the delay variation. By expressing the core as a Partial Integral Equation (PIE) and applying H∞ optimal‑control synthesis, the authors generate state‑feedback gains that keep the system’s ℒ₂‑gain finite for any delay in the interval δmin to δmax.

The method was tested on two benchmark systems. In the first example, the controller keeps the system stable for any delay τ(t)≤2.6 when the minimum delay is zero, improving the previous guarantee of δ(t)≤1. For another variant with a minimum delay of 6.2, stability holds up to τ(t)=6.48, extending the earlier upper bound of 6.35. In the second experiment, the technique secured stability for delays up to 1.16 in a system where earlier results stopped at 1.

The framework rests on known bounds for the delay and on continuous‑time delay trajectories. Its derivation relies on ℓ∞ and ℓ₂ bounds of indicator functions, so the conservatism of the result hinges on those norms. It currently applies only to linear, time‑invariant models with continuous delays; extensions to nonlinear or discontinuous delays are left for future work.