Brancato, Sara Maria (2025) Design and in-vivo validation of learning-based controllers via a sim-to-real approach. [Tesi di dottorato]
![[thumbnail of Thesis_Brancato_Reviewed.pdf]](https://www.fedoa.unina.it/style/images/fileicons/other.png "Thesis_Brancato_Reviewed.pdf") |
Documento PDF
Thesis_Brancato_Reviewed.pdf Visibile a [TBR] Amministratori dell'archivio Download (13MB) |
| Tipologia del documento: | Tesi di dottorato |
|---|---|
| Lingua: | English |
| Titolo: | Design and in-vivo validation of learning-based controllers via a sim-to-real approach |
| Autori: | Autore Email Brancato, Sara Maria saramaria.brancato@unina.it |
| Data: | 1 Novembre 2025 |
| Numero di pagine: | 123 |
| Istituzione: | Università degli Studi di Napoli Federico II |
| Dottorato: | Computational and quantitative biology |
| Ciclo di dottorato: | 38 |
| Coordinatore del Corso di dottorato: | nome email Ceccarelli, Michele [non definito] |
| Tutor: | nome email di Bernardo, Mario [non definito] |
| Data: | 1 Novembre 2025 |
| Numero di pagine: | 123 |
| Parole chiave: | Cybergenetics, Sim-To-Real, Control Based Continuation |
| Settori scientifico-disciplinari del MIUR: | Area 09 - Ingegneria industriale e dell'informazione > ING-INF/05 - Sistemi di elaborazione delle informazioni Area 09 - Ingegneria industriale e dell'informazione > ING-INF/06 - Bioingegneria elettronica e informatica |
| Informazioni aggiuntive: | CICLO DOTTORATO XXXVIII |
| Depositato il: | 07 Apr 2026 06:32 |
| Ultima modifica: | 12 Ago 2026 05:39 |
| URI: | https://www.fedoa.unina.it/id/eprint/17120 |
Abstract
Control strategies are increasingly applied in synthetic biology to achieve robust and predictable cellular behavior. However, the inherent stochasticity, nonlinear- ity, and context-dependence of biological systems make reliable implementation challenging. This thesis develops data-efficient strategies for regulating single cells and microbial communities by combining control theory, machine learning, and experimental platforms. The core contribution is the successful application of a sim-to-real framework for controlling bacterial population density, alongside the design of a dual-chamber bioreactor architecture for regulating population density and composition in microbial consortia. Multiple control strategies were developed, and both the bioreactor architecture and the sim-to-real framework were validated in vivo, demonstrating precise regulation with minimal experimen- tal effort. Additionally, Control-Based Continuation (CBC) was implemented to probe nonlinear dynamics and guide more informative data acquisition. Together, these approaches establish a pathway toward robust, scalable, and data-efficient control of living systems.
Downloads
Downloads per month over past year
Actions (login required)
 |
Modifica documento |
