Particle-filter based vision

This year, we’re trying to approach vision a different way. In the past, we’ve done image segmentation by using various thresholding algorithms and then classified the resulting contours based on size and shape. Our experimental software instead makes numerous guesses of the pose of the object being searched for, and then ranks them by how likely they are to be true, given the images received from the camera (Particle Filtering). The output quickly converges to the true 3D pose of th
Continue Reading →

Move to ROS

View of submarine and pool in RViz
Over the last few months, we’ve been working on moving to ROS (Robot Operating System), an open-source framework for developing software for robots. It includes a message-passing layer that has replaced DDS (Data Distribution System) for us, utilities for managing all the software that the submarine runs, and numerous tools that let us work much more efficiently. In mid-February, the code was mature enough to let us hold position. Since then, we’re been capturing video and sonar data
Continue Reading →

Middle School Outreach

While in Panama City conducting testing in the Gulf of Mexico, Team SubjuGator visited two middle schools in the area. We met with the school’s STEM focused classes including one class that was working on their own underwater platform, a SeaPerch ROV. The visit was focused on robotics at the University of Florida and included a live demonstration of SubjuGator 7 in a very small pool.
Continue Reading →

Testing in the Gulf of Mexico

In support of writing a journal paper with the Nonlinear Controls and Robotics (NCR) group, we conducted controller tests in the Gulf of Mexico off the coast of Panama City, FL. This sea trial tested the robustness of the RISE controller which is implemented on SubjuGator 7. The RISE controller is a cutting edge robust nonlinear controller developed by the NCR group. The testing in the Gulf of Mexico was successful proving that the RISE controller and SubjuGator 7 were able to operate in an open
Continue Reading →

Competition updates and live webcast information.

We’ve been tweeting throughout the week on our twitter account: @subjugatoruf. For the most up to date information about our progress and results, that is the place to look. There will be a live web broadcast of finals on Sunday from 1-4pm PDT via Robosub.org.
Continue Reading →

Packed and Shipped

After a great last week of testing, we’ve been busy getting things ready for the trip to San Diego. First up for SubjuGator 7: a nice bath with Patrick. Well actually, the only shower at the pool was in the locker room, so we had to do what we had to do. Chlorine and anodization don’t get along! Next, we organized our equipment into categories based on utility and daily use. Pre-packing included sorting boxes which stay at the hotel, travel to the Transdec but stay in the car, and ar
Continue Reading →

Mission Testing

Since the leak and thruster failures last week, the team has gotten SubjuGator 7 back up and running. The team has been busy developing and debugging missions for the RoboSub competition. The videos below showcase our current progress on several fronts. Our vision system has been working very well, even in varying outdoor lighting conditions (fitting for the San Diego coast). Hopefully all of our outdoor testing will be advantageous come competition time. The first video demonstrates template id
Continue Reading →

Leaks and failures

The end of June has been rough for SubjuGator 7. Our new pneumatic actuator box managed to form a small leak through our potting material (resulting in ~1cm of water in the box, but no catastrophic electrical failures) and we have now had two Seabotix thrusters completely seize this year (with <12 months use, not surprising given past track records with this brand). Pending replacement cables and potting for the pneumatic housing and the arrival of one additional new thruster, we expect to be
Continue Reading →

Raining Cats and Submarines

We had an unexpected visitor stop by today. Her name is Debby. Tropical Storm Debby, that is. However, she didn’t stop us from testing at the pool. 30 mph winds, drenching rain (the sub didn’t mind), and a few waterlogged engineers can’t stop our motivation. We’re excited to have successfully completed all of our goals for the day, despite the wet inconvenience. As of the time of this post, it has been raining for 14 straight hours. Go Gators!
Continue Reading →

Trajectory Generation

Probably the most visible software change to the sub is our new trajectory generator. It uses a non-linear filter to produce trajectories that are 3rd-order continuous, which our non-linear RISE controller then tracks to produce a smooth moving vehicle. This how our vehicle operated last year with our old trajectory generator, but the new generator brings some new capabilities. Here’s the trajectory it produces for our sub when told to move forward 2 meters, then drive backwards to the ori
Continue Reading →