Explore

Konzept

Embedded System

Input:

Everything from

Testing⁠

⁠

Waterspeed sensor

Hall Effect Sensor (with funnel?), intense calibration necessary to find out the resistance it has

Big version here:

https://blog.arduino.cc/2021/04/30/building-a-low-cost-flow-meter-for-environmental-monitoring-projects/⁠

⁠

GPS (already onboard anyway?)

9 DOF IMU

RC signal (for navigation, manual parameter adjustement, serial comm, & OTA updates?)

power consumption sensor

for both battery and grid power?

underwater noise sensor

load cells (how many kg??)

SD Card

self sensoring (pitch & camber angles, heave, speed on rails...)

angular encoder/ linear encoder

⁠

http://codeforfree.weebly.com/optical-encoders.html⁠

⁠

end position sensor (heaving rod)

IMU in flap

pressure sensor for depth readings (ca. 100hPa per Meter in Water) ✅

Visual methods

Fiducial markers seem to be most effective (better than OptiTrack)

⁠

https://diglib.eg.org/bitstream/handle/10.2312/egve20191277/033-036.pdf⁠

⁠

Camera & Opencv / Apriltags

⁠

https://www.wir-machen-druck.de/aufkleber-din-a4-hoch-210-cm-x-297-cm.html⁠

⁠

MoCon system

Output:

5 Servos

⁠

Sensor Implementation⁠

⁠

Feedback Control⁠

⁠

Reinforcement Learning⁠

⁠

Functionality

operate at different water speeds (and launch from standstill)

Forces on oscillating foils for... paper has info on launch

cornering!

wave harvesting (detect upcoming waves)

Safety stop (e.g. flap entanglement / beaching)?

Felix Tasks

Felix Tasks

Writing

Peripheral implementation

research Waterspeed, Noise, Power consumption sensors

Noise sensors

Control, e.g. Wifi (or SD Card)

Testing

Lake:

Max speed (sinusoidal and asymmetric and step?)

Max speed backwards

Meetings strukturieren

Agenda (incl. Check-In)

Moderation

mitschreiben (incl. Action items)

Test stand

run design by Luca (stand and boat)

CAD

print & build (and CAD wood - load cell adapter)

Reinforcement learning

Build RL algo with mock reward

Implement IO and real reward

test & tune

⁠

https://www.youtube.com/watch?v=6qbW7Ki9NUc

⁠

(youtube video doing similar RL with servos in simulation)

General structure / overview

Using Models (Imdll, Control, RD...), also structure activities with PM

2 way Serial communication

Automated testing improvements

more statistical values

plots with multiple axis descriptions

better filenaming

Visualisation

Serial oscilloscope, e.g.:

https://github.com/dilawar/SerialScope

⁠

or just the typical first search result!!

or universaldashboard?

Powerpoint slides for concept

Using UML

data flow overview

code structure (dashboard, serial)

results & learnings april tags (leave out)

Elektronik ordentlich

Irgendwie überlegen wie das besser geht als immer mitm Breadboard

Code structure

use parameters, modes, documentation...

config files? (JSON or ini?)

ask Luca for style guidelines (and general advice)

April Tags conversion to angle estimation

Calibration,

convert to tangible values (actual % in camber, pitch of cordline, not of servo...),

measure errors,

test with dannys phone for accuracy

Control

State estimator (Lünberg observer for now)

Kalman for Location estimation

Control modeling (Mathematical models, Simulink)

Learnings optimieren

control (position, e.g. with kalman)

Reinforcement learning (different algorithms

plots (do it properly?)

Optional: Easy pulley system

basically just an encoder system, so that we dont measure force, but acceleration / speed of vehicle

Optical testing

use apriltags to plot servo angle vs. camber on a chart?

test how well foils hold up at different frequencies

⁠

This link can't be embedded.

Want to print your doc?
This is not the way.

Try clicking the ⋯ next to your doc name or using a keyboard shortcut (

CtrlP

) instead.