To preface this retrospective, I'd like to explain as to what the Six Degrees of Freedom platform is and what
they tend to be used for. I believe the most common name for them is a Motion Base, a platform that is used
to simulation motion while the things on top of it are in fact sitting still. These kind of platforms are
used in everything from amusement park rides to airplane simulations to space flight training. This makes
the understanding of such technological tools crucial for people training to work on simulations and the
different types of technology based realities.
When we started our first month of the six degrees of freedom platforms, hence forth to be called a Six DoF, we didn't jump straight into building one as I had thought. Instead, the first two weeks of the class was a crash course on everything that the Solidworks design software could do. Once we had done a few test projects and learned to properly measure and replicate an object from the real world into the virtual, we were finally given a single Six DoF and a day, in which we had to measure and measure again before going to Solidworks and replicating the basic outline of the Six DoF. This included taking measurements on the metal poles holding up the top platform and the servos themselves that would be moving the platform, and all three platforms themselves, along with each and every hole that would be used to connect everything together.
These measurements were looked over by the instructor for the course, and once we had everything within a small limit of variation, he gave us the go-ahead for us to take the files to the next class. Once at the next course, we were to begin fabricating all of the parts, using a mixture of tools to craft out platforms. We went through training to understand how to use tools such as the laser cutter, lathe, CNC mill, 3D printers, belt sanders, and a soldering iron.
Starting with the first one, we learned how to properly import a Solidworks model into the launch and control program for the laser cuter, turning the three dimensional model into a two dimensional scan for the laser to trace into the plastic. This included learning what the change the different lines of the scan into, with black lines going to be fully cut through the plastic, and red outlines marking places for the laser to slice away at the plastic bit by bit for an inset cut that left the other side whole. After learning those skills, we turned around and had to understand what we needed to be different from our modeled versions, doing a series of tests to see how far off the laser cutter would change our measurements, as the laser sliced away and made every hole and inset cut slightly larger than we tell it to. With the skills learned, we cut the middle platform for the Six DoF, giving us the first finished piece for the project.
After the laser cutter, we moved on to the three dimensional printers, going through a similar, if slightly shorter, training period. The main thing that was taught for the printer was that it did the opposite of the laser cutter to our models, adding slightly more mass to everything. As such it tightened holes and inset and made extrusions larger, which we had to learn to modify our models that we imported, of which had all three types of styles that would be affected. Once we had the measurements fixed properly, or as properly as we could make it, we went in batches to print the bottom level of our project, and then the top once that was finished, giving us the rest of the platforms for our build.
For the next two tools, we split the group of three students to learn each tool and then switch to the other to make the process faster. This left me learning the lathe first, and the CNC mill after. For the lathe, we had to pick up three other tools to make everything work, which were a hole cutter, a handsaw, and the belt sanders. To start, we were given two eight inch aluminum rods, that were about three inches in diameter, and we needed three three and a half inch rods once we were finished. I measured out where I would need to cut, marked it and locked one half of the rod in a vise attached to the table before cutting through the metal with the handsaw. Once that was done, I took the four pieces with me to the sander along with my calipers to sand off the few quarter inches off to make it the proper length. With few complications, I managed to make all four the right length before taking them back to the table to work on them with the hole cutter, of which I forget the proper name, but it was used to cut the screw ridges into the center of the rods. Doing it by hand was easier than I had expected, but after a few of us took a bit longer than the instructor wanted, the rest were able to use the hand drill to finish their rods off. Once all four rods had their proper holes in either side, I finally went back to the lathe to use an angled bit to cut both the outer edge of the rod to an angled edge, before moving it inwards and doing the same cut to the edge of the drilled hole for the head of the screw to lay snug.
For the final tool, the CNC mill, we had to learn another skill on a program none of us had touched before, which was Eagle, a circuit board design tool. Within the tool, we learned how to place the circuit cuts themselves ,the different through holes for moving the electricity from one side of the board to the other, and the different markings for different types of devices to be attached to the board itself. Once we had the designs up to requirements for our instructor, who required us to not use through-holes and use only one side of the board, we were finally ready to import the file that we had created into the CNC mill's control program. That simply required pulling the file in, and placing the copper board in place on the mill's base plate, locking it down with a line of tap around the outside edge of the material. Once we turned on the external vacuum to suction our board to the plate, we did nothing but sit back and watch as our digital creation was cut into the copper, hoping that everything was prim and proper. After it had finally finished cutting, and all of my nails were chewed out of stress, I was greeted with a wonderfully simple circuit board that I took over to the soldering irons and spent the next day soldering every piece into place. Finally, it was time to assemble all the pieces together.
The final assembly of the Six DoF went much faster than anything else had, with a few people needing to go
back and fix some things that had been overlooked. A few screws here, a few snug attachments to the servos and
finally we were ready to begin working on the coding side for the project. In everyone's free time away from
the class, we had all been making an experience to go along with the Six DoF to show off everything, and now
we had to apply everything together into one, which brought us to our final month of the project, a final 7
day sprint to make sure everything worked properly before showtime.
As can be seen in the video, I decided to make a
mechanized warrior based experience, but since I wished to challenge myself I used flight sim controllers instead
of something more made for a humanoid figure. Once I had everything in place inside of the game-like experience,
then I had to make the platform behave properly and move in time with the mech as it took its lumbering steps.
First came the coding on the platform itself for the servos to work in tandem with one other to achieve the six
degrees of motion that was desired, those being heave, sway, pitch, yaw, roll, and surge. As the servos were
simple rotation devices, attached to another set of aluminum bars, that were then attached to the top plate of
the device, I couldn't just tell them to push the platform forward or roll it, not without code built in place
to explain to each and every servo what its job was. This required going through and manually testing the limits
of each servo and figuring out what I needed each one to do for each desired motion. All in all, the work on
the servos and making them behave took up five of the seven days for the final sprint, but in the end they were
doing just as they needed and I could call for a heave to a certain amount and they would respond doing everything
as required. Once that was finished, it was just a matter of making the calls happen at the same time as the mech
was walking, which may have been easier done should I have been driving my mech with basic controls of forward and
back, but I was using the animations built into the mech to move it around, giving it a more authentic feel.
Finally, the thought occurred to me to use the animations to drive the platform as well, and as such I used tick
marks inside of my animations to send out the position of the mech in its motions which were then fed over to
my Six DoF, moving the platform around in sync for the mech on screen.
For my final thoughts on this project, I feel that while it was rough in some places, I'm content with the way things turned out. Being able to learn so many different pieces of hardware along with learning both new software and new parts of software I already knew, it was a wonderful experience for my brain. If I could go back, the only thing I'd have done any differently would be to put more time in the first month into designing the base platform, as I hadn't realized we had full reign on what it looked like, as long as it still fit the required pieces in. I'm glad to have been able to finish the project, and the knowledge gained from it will help further any work I need to do for any of the hardware used, giving me a starting point to come from instead of being a blank slate.