Why leave UCR UAS?

posted in: uav | 1

If you haven’t heard the news recently, I would like to welcome you to whatever inner circle of confidants I may have.

For the past — three years or so, I’ve been involved in “The UAV project.”  This started out as an American Society of Mechanical Engineers @ UC Riverside (ASME@UCR) project.  The Institute of Electrical and Electronics Engineers @ UC Riverside (IEEE@UCR) began its involvement on the second year of the project when the UCR professional development director noticed a lack of progress despite a generous donation from Northrop Grumman.

My involvement began from the IEEE-side-of-things (2012-2013).  I was a Freshman at the time, and I wanted to make an impact on UCR’s public appeal.   I wasn’t accepted into many colleges; my GPA was low and my SAT scores were okay.  (3.12, 2230 (780 reading, 780 math, 670 writing))  But I really enjoyed the UCR campus.  It’s the strange “something feels right” that you get walking through parts of UCR.

But beside from that, there were conflicts with the ASME and IEEE portions of the project.  Neither side really liked the other, and this was created by tension between the coordinators from both sides.  The atmosphere around the project was grim and little work was being accomplished.  So I left.

The year afterwards, a good friend of mine took over the project.  Of course, I didn’t know that he would be a good friend of mine at the time.  Russell was the IEEE@UCR Project Chair from 2013-2014.  His duties included watching after the project (the UAV stuff), and he took on the entire thing with a bit of anxiety.  We knew little if nothing about UAV’s.

The same year, I remained on the IEEE@UCR board as Micromouse Coordinator.  Micromouse is a competition in which students are tasked with building a robotic mouse that can solve a 16×16 maze to reach the “cheese” in the center.  I was the only person on the board at the time who would do the job, but I can’t say that I wanted to.  The previous person who did it was more or less removed from the IEEE@UCR board due to internal conflicts after the failed UCR Micromouse Competition.  I helped with the micromouse maze construction as much as I could, but there wasn’t enough time to complete it.  I also helped with spec-ing the award plaques that would be handed out to the winners of our micromouse competition, and kept the third place plaque when it was given to me in the aftermath.

3rd Place Micromouse

It is interesting to note that I also met another good friend of mine, Kevin, during micromouse things.  Kevin would probably have won second if we had the competition.  His involvement with the IEEE began after my Freshman year.  He became an officer this year (2014-2015).

Being Micromouse Coordinator didn’t go very well for me.  Perhaps it was telling people that the project would be hard.  Maybe it was the fact that my Micromouse did not perform well and I did not believe in my abilities to any extent.  Maybe I was forced to do it.  Maybe it could be my PI’s words of discontentment about Micromouse and its importance in research.  Maybe it was the lack of advertisement throughout the courses.  Who knows?  I had few if any people showing up for Micromouse workshops, and I didn’t have time to prepare for Micromouse workshops.

At the same time, I wasn’t doing very well in my classes.  CS 111 (Discrete Structures) didn’t capture my attention (I would later on go to pass the class with a B-.)  EE 001B (Engineering Circuit Analysis II) presented information way too quickly and I did not absorb much of it at all (I am currently retaking the course.)

For whatever reason, I was drawn back into the UAV project by things that were left undone or weren’t done correctly.  Day after day, I would find myself doing things my way, the slightly better way.  Lots of my found faults with the onboard computer would work.  My proposed solutions to in-air disconnects and image transfer were done.  Unfortunately, my large influence in the matter rubbed a (now) good friend of mine quite a different way.

Brian was under the impression that I wanted to replace him in the project.  I had no ill-intentions, and I was motivated by the mere idea of “doing things better.”  It could have been my demeanor to the idea of “You have merely adopted Linux.  I was born in it — molded by it.  It wasn’t until college that I’ve seen Windows running on ’embedded’ platforms.”  Or it could have been the fact that I was touching a lot of his stuff and doing things the way that I would do it.  Fortunately or unfortunately, I’m a jerk when it comes to implementation.  I should mention that our relationship has been patched up.  Part of it began with my creation of a “good list” and a “shit list” which I will mention later.

(maybe put a pi here)

I didn’t have a big role in the mess, but I was very happy just doing the communication between the camera and the groundstation.  But of course, there were more things that had to be done.  For one, we had our image detection running on the ground station.  However, it would take far too long to process one image to process the whole batch of images that had to be processed.  I was tasked with programming a multi-platform system to interface with MATLAB and process through every image.  For some reason, I had decided that using Boost libraries was the way to go.

I would spend far too long trying to figure out Boost and get nowhere.  (I now know that I like low-level C for some reason.  The lower-level, the more fun I have.  It’s no surprise that embedded C is a pleasure for me.)  The fact is that I would be essentially writing a supercomputer from scratch.  And this was to be made two weeks before the competition.

It eventually came time for the UAV competition.  Our hard work would come to fruition.  Our test flights would become preparation for the competition.

We really didn’t know much about how things would be done at the competition.  Even though it was a lot of fun, we were trying out new high-torque servos on our airframe and the entire system would brown out.  Life was hell for a bit when we were sorting out the brown-out issues.  But we flew and actually took aerial photos of many “targets” that we are supposed to photograph.  If I had written the multiplatform supercomputer-from-scratch, it would not have been very nice anyway.  The person in charge of image detection broke his code beforehand and did not have a version that worked.

(insert UAV successphoto here)

Everybody had a positive attitude about our try at the competition.  We honestly didn’t think we would make it for a short time.  We won twenty-fourth place.

Although I did quite poorly and was placed on academic probation, I was somehow allowed to be the IEEE@UCR projects coordinator.  I was tasked with presenting at least one interesting project for people in the IEEE@UCR to do per quarter.  This was a pretty rewarding position to have.  I did numerous project presentations such as the IEEE@UCR Soldering Oven and the Yellow-Brick-Laptop-Charger.  Apparently, I gave someone the kick-in-the-pants to begin his involvement in the IEEE@UCR and learn about the exciting world of embedded Linux devices.

The person who broke his code and had passed me the task of supercomputer-programming became UAV coordinator.  I was noticeably upset by this and had thoughts of quitting, but I became the communications team lead.

The summer before the school year began, I opposed our decision to go with a machine vision camera that would cost us near $2,000.  It was a very simple opposition on my behalf.  $2,000 is a lot of money and we don’t know about how the software will be for the camera.  Upon further review of the camera performance metrics, I also thought that it would be a bit slow to capture images.

I had other things to worry about.  At that time, I decided to spearhead the development of better communications software.  It is known that rsync [link here] was not designed for continuous synchronization, but was actually designed for a few synchronizations at a time.  So I programmed BISON-Transfer and BISON-Transferd [link this](the d is for daemon per traditional *nix naming convention).  I had the idea that it must be reliable stuck in my head, so I made it as reliable as I possibly could.

I also had the idea of making other software that would integrate more UAV control into the hands of the operator.  This would be a multi-user software system that would allow a multiple users to interact with camera picture taking, communications software, and telemetry features.  Unfortunately, the breadth of this required a robust (decidedly custom) communication protocol and multiple servers to be developed.  The person assigned to this was Cody Simons who tried his best (and frankly did very well) developing this software.  It was not “production-ready” by the time we launched the UAV at the competition.

We had created an airframe called “the Bison.”  The name, of course, comes from our “Spirit Bison” that was found on the airfield before our first competition.  We eventually concluded that it would be good luck to have the bison with us wherever we went.  This was largely engineering superstition, but it somehow worked until somebody decided to use the idea of “Spirit Bison” beyond an intended line.  This would stretch to making an account on Facebook called Spirit Bison and denouncing the ASME portion of the project.  Unfortunately, the airframe and the Spirit Bison were not production-ready.

(spirit bison picture)

Other things that weren’t production-ready were the camera and picture location.  The camera lens had suffered a crash a week beforehand and its focus ring could not be turned, so focus was adjusted using the thread onto the body.  Even if it had been adjusted correctly, the exposure settings and gimbal could not keep up with the movement of the UAV.  The lens also had a focal length that made target tracking quite difficult.

In all, the team had a very negative appeal to the entire competition for me.  The car rides that were lighthearted bickering and friendly teasing became actual teasing and unfriendly bickering.  Somehow, our mission performance was eighth in the competition and we ranked tenth in the overall rankings.

It was at the awards banquet when I really started to consider whether tenth place was worth my pain, and I realized at the awards banquet that it was not.  Of course, part of me wouldn’t accept this consideration and I eventually turned “no, it is not worth my effort” into “yes, it is worth my effort” by the time I got back from Maryland.

I spent the Summer without an internship and found something to take up my time.  It’s for the better, of course.  I spent most of my summer recovering from the ill-treatment received and preparing to be UAV coordinator.

My first meeting was more or less a no-show with everyone.  Very few people showed up to our internet-meeting despite the effort that I put in to it.

My second meeting as UAV coordinator, I somewhat rejected the camera that has been failing us.  The previous UAV coordinator seemed to have a mental meltdown about the whole ordeal, and we decided to keep the camera with the claim that “it works now, we can take a foamie and fly it tomorrow if you want.”  Me being me, I kept a straight face and calmly spoke to everybody else.  The previous coordinator, of course, took control of the rest of the meeting and asked questions out of the order from the meeting agenda.  The last thing I want is to have to drag someone out of a meeting, so I kept with it.

The summer also involved the moving of many too many things to our new room at CE-CERT.  We were led to believe that our room was to be immediately used to house faculty for the Bourns Hall B renovation.  So we packed up all of our things from the old “UAV room” and moved to CE-CERT with vans.  At the end of the entire thing, our president sent an email marking our official move-out.  As of this day, no faculty are in Winston Chung Hall 233.  CE-CERT remains our storage area instead of a dedicated workshop for UAV things.

CE-CERT also holds the materials for our indiegogo rewards.  The Indiegogo failed to make its goal, but the money was accepted and the rewards have to be handed out.  However, the person in charge of it has vanished from UCR UAS.  He has, after quite a long time, given us the production designs for making a 3d printed plane.  I gave his late designs to the 3d printer manager who rejected them.

And after more internal conflicts, it no longer became worth it to put up with the project any more.  Maybe I should have just gone with the knee-jerk reaction of “it’s not worth it?”

Leave a Reply

Your email address will not be published. Required fields are marked *