Erdmann, Niklas (2022) Dynamic Modeling of a P(VDF-TrFE-CTFE)-Composite Soft Actuator. Master's Thesis / Essay, Artificial Intelligence.
|
Text
mAI_2022_ErdmannN.pdf Download (5MB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (124kB) |
Abstract
Soft actuators, in contrast to traditional actuators, are made from soft and compliant materials. A category of soft materials are electro-active, showing a response to an applied electric field. Electrostrictive materials, a type of electro-active material, have a quadratic response between the applied electric field and the strain they produce. This project attempts to model an electro-active soft actuator, composed of an active layer of electrospun aligned nanofibers made from the P(VDF-TrFE-CTFE) electrostrictive polymer integrated in a PDMS silicone matrix. The actuator features a passive, Kapton layer and electrodes made from a conductive carbon powder silicone mixture. The soft actuator shows complex dynamic responses to electric field stimuli and natural individual differences between samples. These complications motivate a novel dynamical modeling attempt, employing an Echo State Network. The trained model, tested on new data, displays a normalized root mean square error (non-normalized in parenthesis) of 0.433 (0.106 mm) and 0.280 (0.019 mN) towards modeling tip deflection and blocking force exertion respectively. To show applicability and transferability of the model, the blocking force network is tested in a more functional context: A two actuator gripper. Despite noisy task and observational settings, the modeling and the gripper experiment suggest Echo State Networks work very well with electrostrictive soft actuators.
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Supervisor name: | Carloni, R. and Jaeger, H. and D\'Anniballe, R. |
| Degree programme: | Artificial Intelligence |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 11 Jul 2022 10:25 |
| Last Modified: | 11 Jul 2022 10:25 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/27734 |
Actions (login required)
 |
View Item |