Paun, Gabriel-Vlad (2022) Passive hydrodynamic imaging using the pressure amplitude and flow velocity. Bachelor's Thesis, Artificial Intelligence.
|
Text
ReworkedVersionS3811816-3.pdf Download (576kB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (121kB) |
Abstract
Lateral lines are sensor arrays designed to detect objects moving through fluids. They possess distinct advantages compared to other types of sensors, being undetectable, usable in low light conditions and suitable for short-range detection. This study investigated whether the addition of hydrodynamic pressure sensors to a velocity-sensitive 2D lateral line can improve the accuracy of object localisation. An extreme learning machine (ELM) was trained on simulated data generated by placing a vibrating sphere along different grid points on a 2D map. Multiple configurations of lateral lines were tested, by gradually increasing the number of pressure sensors and decreasing the number of velocity sensors, and gradually pulling the y sensors away from the x sensors. The ELM learned to output the position and direction of vibration of the simulated sphere. Two different methods of positioning the velocity sensors relative to the pressure sensors were compared, one where both are located along a single horizontal line, and one where the velocity sensors are placed on top of the pressure sensors, on a vertical line, forming the shape of an upside-down T. It was found that pressure sensors can be used to replace some of the velocity sensors is a 2D lateral line; for determining the position of an object, few velocity sensors are required. However, determining the direction of vibration was seen to require at least as many velocity sensors as pressure sensors.
| Item Type: | Thesis (Bachelor's Thesis) |
|---|---|
| Supervisor name: | Netten, S.M. van |
| Degree programme: | Artificial Intelligence |
| Thesis type: | Bachelor's Thesis |
| Language: | English |
| Date Deposited: | 07 Oct 2022 13:51 |
| Last Modified: | 07 Oct 2022 13:51 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/28817 |
Actions (login required)
 |
View Item |