Parley SnotBot Alaska, Take Two
I am writing this email en route to Alaska. Yes, we are heading back up to Alaska for the second time this year, and it is exciting for many reasons. The first of which being that we are continuing our Parley SnotBot collaboration with INTEL and Ted Willke’s team at the Mind’s Eyes lab.
On the scientific front, to be able to go back and study the same population later in the same year is not something that we get an opportunity to do that often; couple this with the advanced tech the Mind’s Eye lab are bringing to the table, and we have an exciting expedition ahead of us. We are returning to the area we worked this summer in the hope of seeing some familiar whales and documenting how their behavior, interactions, and body condition have changed since we last saw them.
Since the last expedition Ted, Bryn, and Javier have been working like crazy to expand their fluke ID database and further develop their algorithms so that they can better identify individual animals and better estimate size/girth with their volumetrics work (more on that later). The weather is likely to be a challenge, but we have stacked the odds as far as we can in our favor. We are of course working with the Alaska Whale Foundation and Dr. Fred Sharpe, who should keep us pointed in the right direction.
We will be staying on a boat instead of with our good friends at the Kake Kwaan Lodge (sorry Tinker) so Internet is going to be tricky, but my goal is to send you expedition blogs whenever possible. We do have expedition cameraman extraordinaire Christian Miller with us, so expect some pretty dramatic photographs – I have attached a few from the summer expedition.
I remain an avid proponent of drones for whale research but also of drones for ocean research. I think that these machines are going to permeate every aspect of ocean research, changing the way we do our work and hopefully resulting in a democratization of science, with more people collecting more data in more locations, more affordably, and more safely. It’s just amazing to me how far this tech has come in just three years and how far it has permeated (and will permeate) so many aspects of our work. Not just the drones and sensor packages but also the support equipment. I was in our robotics lab last weekend 3D printing small mounts to hold petri dishes on the INTEL Falcon 8 drone for this expedition (photos attached) and I realized how I was already taking this amazing affordable tech ($2,500 printer) for granted. Yes, one of the students (Austin, pictured below) had to do the design work for me (10 min?) and then we were printing out custom mounts for two different drones just 30 minutes after coming up with a design concept – crazy.
And don’t even get me started on the Artificial Intelligence and machine learning work that Ted and his team are doing. I bring this up because while I believe that the work we are doing now is innovative and, dare I say, disruptive (in a positive way), I am excited by what we will likely be doing with drones in two years from now. I will admit to finding the challenge of trying to predict where this tech is going and how we will make best use of it very thought-provoking.
As an example, when we are collecting snot we have to have our Parley SnotBot at the right height above the whale, we have to be the right distance behind the whale’s blowhole (varies due to the speed of the whale) and we have to be at the right angle behind the blowhole (depending on crosswind) so that we are in the best position to collect the largest amount of snot that we can. It’s not as easy as you might think. I have thousands of hours of flying time, which most people don’t have – making the current Parley SnotBot tool not as replicable as we hope for. That said, I believe that in a year’s time (two at most) pilots like me will just be overseeing the snot collection flight with the SnotBot auto sensing drone (version 22 🙂 analyzing all of the above parameters in real time and far more accurately and consistently than a human can – consequently bringing in more and better data and bringing this tech into the hands of pilots who have training but perhaps not as much as our pilots have and need now.
From a wet and windy in Alaska.