Friday, 20 June 2014

Trixie's Success

This years competition has now come to an end with MUN SailBOT placing 3rd (a close 2nd) of 5. Trixie performed exceptionally well this year and we were all pleased with her success. Many lessons were learned during the event and new ideas for the design of our next vessel are forming.

A special thanks to our sponsors for making this success possible.

Friday, 6 June 2014

Its The Final Countdown

Well a lot has happened within the last month or two and its been quite a successful time. After a hard 10 months of work, the beautiful Trixie has now been completely designed and built and as of a couple of weeks ago, the testing stage has begun. But now the time has come to stand tall and really show off Trixie's skills during the annual SailBOT competition which is held in San Francisco this year. Very soon, we (6 team members) will be leaving together with Trixie in our pocket and hope in our hearts!





Saturday, 10 May 2014

Little Dragon Gets Smart

Over the past few months Zhi Li has been working on a new test vessel.  Something small and easy to deploy, while at the same time being capable of mimicking the 2-m vessel Trixie such that any code developed can be transferred over to the larger vessel. In essence, a navigation code development platform.

The boat picture below is the Little Dragon, a 1-m RC vessel which has been modified to work with the Xbee radio system, GPS and wind sensor used on Trixie.  The two sensors  relay information to a BeagleBone nested inside the vessel.  The Beaglebone then relays this information through the Xbee back to the shore based computer. 

The beauty of this system is in how the autonomous code is integrated.  Instead of being downloaded onto the Beaglebone, the code is run wirelessly from the shore computer.  For testing and trouble shooting navigation code, this allows changes to be made rapidly without having to bring Little Dragon back to the dock.

Little Dragon should be a great asset for this year, as well as future generations of the team, having successfully allowed the build of the vessel to be carried out independent of the testing of navigation code.

Below is the code that runs the communication and control system.  Open to anyone and every one to use. 

Code Link:
https://drive.google.com/file/d/0Bwp9qS8JICuZMENNNGVScUptcGM/edit?usp=sharing






Thursday, 24 April 2014

To San Fran, with the wind at our back!



Progress Report: our hull is well underway! We started work on it back in March during our prototyping phase. The photo below shows a half-scale version, constructed with cardboard stations and household insulation. The half model was built to familiarize the team with an entire cycle of the build process. The prototype allowed us to test ideas and construction methods without having to put in large amounts of time or money (Neither of which we have large amounts of). Some novel ideas were tested to great success; such as using saran wrap as a mould release agent.  The half-scale model also identified the number of stations that are needed to fully define the hull form, and gave us an idea of how much work is required to create a fair final product.  The team also discovered the tricks and problems of working with foam; mainly with regards to how soft it is.
Half-scale prototype – wrapped in saran wrap and ready for glass.
Following on the success of the half-scale, the team began building the full-scale version. For this, we started with laser-cut 1/8" birch stations, profile and deck. The profile was added due to the size of the hull, the thinking being that another defined line would help with the sanding process. Below, Adam Day can be seen putting the "skeleton" of the plug together. The deck had some warp in it and had to be held down with drafting ducks from the senior naval arch. class room (better use for them than door stops...) while the putty that was used to assemble the skeleton cured.
Adam Day and the "skeleton".

Skeleton puttied, loaded down with ducks.
Once cured, the next step was filling the skeleton with foam. This took much longer than expected, more time than was needed on the half-scale model. Each piece had to be cut square and some sections had to be custom fitted to length. The final result looks a little rough around the edges but with some sanding should shape up nicely.
Adam Day and Martha Hermann sizing up the foam.
  

The team is aiming for an early launch, so expect more posts to follow as the hull emerges from the foam! 
Now... We Sand!

** All photos taken by Emma Williams**













Saturday, 8 March 2014

A Thank You to Trapper John's!


MUN SailBOT would like to thank Trapper John's for hosting us for our fundraiser on Saturday, March 1, 2014. In total we made over $250 at the event and a good time was had by all. The money will be going towards building our new boat as well as the competition costs.

Saturday, 22 February 2014

A New Member of the Family

As of early last week, the MUN SailBOT team has gained a new ally in their quest to solve science once and for all. 
Ladies and gentlemen... Petunia the Pump!



Petunia is a classy pump come up to help all the way from California, USA. She has thus far been a great aid with the development of some components for the new vessel Trixie. These components include a new rudder, a hatch prototype and a test section of the keel. There are sure to be many other items that Petunia will generously lend a hand with - and we look forward to each one!


Left to Right: Petunia and Liam Johnston





Monday, 27 January 2014

"If you can't fix it with a hammer, you've got an electrical problem"



Well it looks like we've got an electrical problem then.  After a long spell of debugging communication issues between the Arduino and the BeagleBone, we had run into another challenge: the GPS.
Pictured above the team has taken the Arctica (the boat) and her sensors outdoors to see if we can get a fix with the GPS.  The GPS is the eyes of Arctica and without it  she would loose all sense of purpose in her life. At the time we couldn't get the GPS to pick up any satellites and thus couldn't get a fix. Either the GPS itself (hardware) or somewhere within the many lines of code that Daniel has hand crafted (software), something wasn't working.

After some background research on the GPS and how it works, along with a new GPS, we managed to get a proper fix.  However nothing is ever easy and we are now faced with the task of sending the way points from the Arduino to the BeagleBone for course plotting.  This involves setting up the "mailing procedure" of the data. This procedure lets the BeagleBone know what it should be receiving (GPS coordinates) and tells it not to sign for the parcel if it thinks it's getting garbage data.

Guess we'll be giving the hammer a rest for a bit. Although some days I'm sure Dan feels that a little persuasion here and there wouldn't hurt.