Milestone 1
Method for Wave Height & Frequency (Brendon and Jasmine)
Method for Dead Reckoning (Samprith)
Method for Wave Direction (Howard)
To get a B in this class, I need to collect data from an experiment where we float a surfboard with a Smartfin next to CDIP Buoy #201 (closest buoy to Scripps pier), analyze the data, and write-up the results of how our two methods for determining wave statistics--spectral analysis and wave train analysis to determine wave height and wave period--compares to CDIP results from same day data. To get an A, in addition to the above goal, I also need to analyze data from an actual labelled surf session and do a similar write-up which compares our methodology and the Smartfin’s results to same-day CDIP buoy data for wave height and wave period.
To get a B, I must first use Buoy-Calibrated Smartfin data to determine wave direction, so that we can derive a directional plot using polar coordinates, as well as a 3-D plot of Azimuth and Altitude (this is assuming the Smartfin is oriented on the buoy so that it is riding with the wave and not against it). To get an A, must compare results against actual CDIP buoy data and get less than 10% error, furthermore, need to make method work for real surf data, and differentiate findings based on different surfer behaviour, such as riding against the wave, riding with the wave, changing direction, wiping out, etc.
To get a B in this class, I need to write up code to determine the initial orientation of the Smartfin and analyze the results of the code run on controlled data sets. To get an A, in addition to the above goal, I also need to analyze data from an actual surf session and do a write-up which compares how accurate the orientation code is on real-world data compared to controlled data, and document points of interest in the code where miscalculations may occur. I also need to create a simulation to visualize the differences between what we expect to occur, and what actually occurs.
To get a B in this class, I need to create a basic implementation to determine surfer dead-reckoning. To get an A in the class, in addition to the above goal, I would like to build a more robust dead-reckoning model that predicts movement to a high accuracy using orientation rotation matrices, and gyroscope and accelerometer readings (although this will depend on brendon's orientation results), and I intend to incorporate orientation into the our existing methodology for calculating significant wave height in order to create a dynamic model that we be able to ingest real world wave data (although this will again depend on brendon's orientation code).
Created by: Howard Wang, 05/08/19