Linear dynamic models are extremely useful in autonomous vehicle controller development because they are straightforward to estimate from real data and they enable advanced control system design. To date an in depth investigation of the possibility of using a linear dynamic system model to represent sailing yacht dynamics has not been made. A non-linear simulation of a 10m racer/cruiser racer and full-scale data collected aboard a 23 ft day-sailer are used in conjunction with the one-shot least squares estimation algorithm to evaluate this possibility. Using the non-linear simulation data it is found that it is possible to use a linear model to represent sailing yacht. Full-scale data collected while sailing upwind on a port tack is used to estimate 10 different linear models.
Least Squares Estimation of Sailing Yacht Dynamics from Full-Scale Sailing Data
Legursky, Katrina. "Least Squares Estimation of Sailing Yacht Dynamics from Full-Scale Sailing Data" Paper presented at the SNAME 21st Chesapeake Sailing Yacht Symposium, Annapolis, Maryland, USA, March 2013. doi: https://doi.org/10.5957/CSYS-2013-002
Download citation file: