Router 1

I’ve not blogged in over six months, other than the mermaids, who are on autopilot.  This is for a variety of reasons, but for archival purposes I will try to get some of my racing stuff up, though this season has not been one of distinction.

In the face of perennial crew difficulties, Annika and I are pretty much planning on doing PacCup 2012 double-handed, in the ultimate daddy-daughter adventure.   She will be up from school for the winter holidays, and we are going to take our First Aid course together and if we can get reasonable weather, an overnight offshore cruise.

Meanwhile, I have been amusing myself by writing my own routing program in Mathematica, since that is what I know, and it supports all sorts of graphics and data-juggling features, and even can directly import GRIB files.  The pic above shows a test run to determine the leaders (outlined in red) of a  fleet of isochronic boats.   I am working on both the standard isochron method, and a stretchy-string idea which may be my own, or maybe not.  It’s a pretty difficult problem either way.   I’m using some polars I found for a Beneteau 36.7, and the program still chokes when you are sailing along through the non-stationary wind field and you go into irons for too long.    I’m not sure how to sensibly insert short-tacking into the search method, other than by Monte Carlo’ing the heading and dealing with a vast  number of branching possible routes.   Fortunately, the SF to Kaneohe route I care about has no land to dodge around, so I can ignore collisions.

4 Responses to “Router 1”


  1. 1 Kelly

    Sounds like an amazing adventure is about to happen. Hope all goes well and you have fun. Don’t forget to take pictures of this, and please post them. 🙂

  2. 2 Buck

    As a lifelong mainframe (business apps) programmer I find this programming project an exquisite challenge. It smells very difficult. If you set the interval to an hour and there’s a nice bit of wind that’s an hour and a half’s sail away, you may be unable to reach it at the next interval. Set the interval too low and the calculation burden gets insane.

    Thanks for the links; I had no idea someone was trying to solve this with software. Some very, very interesting reading.

    If I were sailing double-handed I’m not sure I’d want to do a lot of short tacking. Maybe an alternative of a pair of tacks making up half the current remaining interval? Sorry, it’s such a different problem from what I usually deal with and therefore irresistible! I look forward to hearing more, as you make progress with it.

    Best of luck on your plans – oh how I wish I were on the West Coast!

  3. 3 Blur

    First of all, congratulations to decision to go doublehanded. A fantastic opportunity to spend time with people you love.

    Then, when it comes to routing, it’s a challenge. And the more you learn the more complex it gets 🙂 We’re using Expedition thats pretty impressive. Even if you write your own, be sure to dowload a 30 day demo to look at the features and workflows.

    For me the key is to be able to dou several routings and then compare. Different models but also move them back and forth in time (the model is right, but the timeing must be adjusted).

    Also, here’s a pretty cool project on how to do multiple-route comparisons: http://www.blur.se/2011/03/24/weather-routing-ensembles/

  4. 4 bastcilkdoptb

    I simply wished to thank you so much once more. I am not sure the things that I could possibly have gone through without the advice discussed by you directly on such a area of interest. This was a troublesome setting for me personally, nevertheless witnessing the very professional fashion you treated it forced me to cry for fulfillment. Now i am happy for the support and even expect you recognize what a powerful job your are accomplishing training most people all through your website. Most likely you haven’t met any of us.

Leave a Reply