Policy Optimization with Differentiable MPC: Convergence Analysis under Uncertainty
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911592762834944 |
|---|---|
| author | Zuliani, Riccardo Balta, Efe C. Lygeros, John |
| author_facet | Zuliani, Riccardo Balta, Efe C. Lygeros, John |
| contents | Model-based policy optimization is a well-established framework for designing reliable and high-performance controllers across a wide range of control applications. Recently, this approach has been extended to model predictive control policies, where explicit dynamical models are embedded within the control law. However, the performance of the resulting controllers, and the convergence of the associated optimization algorithms, critically depends on the accuracy of the models. In this paper, we demonstrate that combining gradient-based policy optimization with recursive system identification ensures convergence to an optimal controller design and showcase our finding in several control examples. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_01940 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Policy Optimization with Differentiable MPC: Convergence Analysis under Uncertainty Zuliani, Riccardo Balta, Efe C. Lygeros, John Systems and Control Optimization and Control Model-based policy optimization is a well-established framework for designing reliable and high-performance controllers across a wide range of control applications. Recently, this approach has been extended to model predictive control policies, where explicit dynamical models are embedded within the control law. However, the performance of the resulting controllers, and the convergence of the associated optimization algorithms, critically depends on the accuracy of the models. In this paper, we demonstrate that combining gradient-based policy optimization with recursive system identification ensures convergence to an optimal controller design and showcase our finding in several control examples. |
| title | Policy Optimization with Differentiable MPC: Convergence Analysis under Uncertainty |
| topic | Systems and Control Optimization and Control |
| url | https://arxiv.org/abs/2601.01940 |