Metraipolis › archive › dc_0dff892eb48e

Deep Reinforcement Learning with Smith-Predictor Delay Compensation — arXiv:2611.00668

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.00668
License (as archived): CC0-1.0
Archived: 2026-11-03T11:54:43.363Z
SHA-256 of the archived text: a16d3bd20f69d6440935ae46bf6d3480e292f31bf25326b576b035a4fcfc9f63
Stable URL of this page: https://metraipolis.com/documento/dc_0dff892eb48e

How to cite

Deep Reinforcement Learning with Smith-Predictor Delay Compensation — arXiv:2611.00668 — arXiv (en), https://arxiv.org/abs/2611.00668 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0dff892eb48e (Metraipolis archive dc_0dff892eb48e, sha256 a16d3bd20f69d6440935ae46bf6d3480e292f31bf25326b576b035a4fcfc9f63)

BibTeX

@misc{dc0dff892eb48e,
  title = {Deep Reinforcement Learning with Smith-Predictor Delay Compensation — arXiv:2611.00668},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00668},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0dff892eb48e; sha256 a16d3bd20f69d6440935ae46bf6d3480e292f31bf25326b576b035a4fcfc9f63},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00668. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Time delays can degrade feedback performance and destabilize a closed loop that is stable without delay. This paper studies Smith-predictor-assisted reinforcement learning for systems with constant input delays. A deep deterministic policy gradient (DDPG) actor is trained on a nominal delay-free environment and deployed with a Smith predictor using a fixed nominal delay. For linear controllers, we derive exact closed-loop maps under plant and delay mismatch and provide small-gain and $\nu$-gap robustness conditions. A separate delay-independent Razumikhin condition gives a sufficient stability test for a fixed Lipschitz state-feedback policy. We also study output-error delay identification from a fixed input--output record, deriving the loss gradient and establishing projected-flow descent and local exponential convergence under exact model matching and an excitation condition. These identification results do not imply stability of an online adaptive predictor. Numerical experiments retained from the conference study compare PID, LQR, and DDPG controllers on stable and open-loop unstable first-order plants with valve stiction. The results illustrate the benefit of predictor compensation in the tested configurations and the sensitivity of standalone DDPG to delay and plant changes.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv eess.SY, nuovi lavori del giorno).
Categorie arXiv: Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2611.00668 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T11:15:22.662Z dalla pagina https://arxiv.org/list/eess.SY/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).