I had prioritized the ECE capstone in the previous weeks knowing I would be out of town this weekend and not have the time to spare. While I have not had as much time to do more tangible efforts, my spare time amidst my other obligations was often captured by design considerations or our individual carom bots and researching how others have done similar ideas. I considered this okay for now, and didn’t redelegate as I wanted to take time to really think things through before locking a final design in anyway.
FreeCAD is a software I am learning how to use to make our 3D printed housings and designs. I as the robotics major am most familiar with design, and have given semi-detailed plans for a mechy to cad, but before this I have not done much CAD myself beyond intro to Mechy freshman year. I am therefore going through tutorials to develop that skill as I think about our design in parallel.
What is the priority of the design? My primary consideration was in creating something with agility and speed. Agility to navigate rough terrain, and speed to match the urgency of a search and rescue operation. It is also necessary to consider how either the robot will come back to the user, or how can we make the system so cheap we don’t care if some or even most can’t come back.
The design from our presentation last week (I can’t get the image working through my iPad, sorry, see the team status report post) is the base design I arrived at in the previous weeks, from which I am now refining. The rest of this post will give a brief look at what I have considered.
To enhance agility, I am considering incorporation of a spring / launching mechanism in the arms, where they store energy and release it to “jump”. (See https://youtu.be/daaDuC1kbds?si=DPnzXT5Hto2RL203 and https://youtu.be/xUUW6SYl_ak?si=parwDyRLWtX7nn9E for the general idea)
Our wheels to enhance speed do not need to be on the end. They instead could be around the middle of the arm. This leaves room for more easily attaching our catoms together, as now the ends are freed up again. It could also serve as padding for the catom similar to a door stop if the launch mechanism is incorporated.
The method of attaching catom to catom doesn’t need to use magnetic actuation. I thought at first that at the very least it will need a complementary connection mechanism. However, for our use case I don’t think so if we consider the alternatives. 1) a latching/coupling mechanism would require a level of precision that our catoms don’t have. It also would need a motor that either is on the end (which makes it harder to move the end effectors around) or it needs a shaft that runs through the arm. This is different than a magnet that may take similar space, as the shaft would be moving and potentially grinding on other components we want in the arms if we go the route of adding sensors to them. 2) if we don’t have a latch/coupling mechanism, then it will not be stable by default, and will require some constant force to keep any connection in place.
This is why after deeper consideration I still think a magnet is great, as it by default will be attracting the ends together to maintain connection, and the magnet doesn’t require movement to disconnect. This allows us to maintain low power, and smaller form factor as there are less parts to worry about, and it also gives it a larger life cycle since it isn’t wearing internals out over time.
Once these design decisions are finalized, then we can consider stuff like battery life and servo strength requirements. I have secured servos from robo club already to test, and coincells we can potentially use. I want to stress test servos running off a LiPo coin cell to see how they can handle a load. I also secured a board for Brooks to start running software tests on.
Finalizing the physical design by next week is my goal, but at the very least we need a finished BOM to start developing a PCB board to order. This means it may be okay to not have the design in CAD, but the decisions that lie as dependencies for our battery life and servos must be done by next week (more like Wednesday honestly) so we can have the BOM done next week.