Tajgler, Veronika (2022) Control of nonlinear fully actuated physical system via Neural Interconnection and Damping Assignment - Passivity Based Control. Integration Project, Industrial Engineering and Management.
|
Text
TajglerV_s3351068.pdf Download (27MB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (122kB) |
Abstract
Recently developed Neural Interconnection and Damping Assignment - Passivity Based Control (IDA-PBC) can be applied to a wide range of mechanical systems described in the port-Hamiltonian (pH) framework. It demonstrates a great potential to overcome an obstacle that comprises solving the matching equations of closed-loop system. This control scheme relies upon exploiting the universal function approximation property of neural networks and reformulating non-parameterized IDA-PBC into a supervised learning optimization problem. This study aims to highlight the applicability of Neural IDA-PBC by conducting real-life experiments on a fully-actuated physical system, namely working with a swing-up pendulum problem. Finally, the performance of the Neural nonlinear controller is compared to the analytical IDA-PBC, and the system’s transient response to different controller parameters is analysed.
| Item Type: | Thesis (Integration Project) |
|---|---|
| Supervisor name: | Jayawardhana, B. and Munoz Arias, M. |
| Degree programme: | Industrial Engineering and Management |
| Thesis type: | Integration Project |
| Language: | English |
| Date Deposited: | 15 Jul 2022 14:49 |
| Last Modified: | 15 Jul 2022 14:49 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/27935 |
Actions (login required)
 |
View Item |