Robot Arm C

[harryk]

That reminds me. I used to recommend lithium polymer batteries, because they're getting cheaper and can supply enough current for almost anything. I don't recommend them anymore, if you're the kinda person who makes stupid mistakes like me. Because sometimes you forget to balance them. And then sometimes when you charge them one of the cells gets overcharged. And then the whole pack catches on fire so you have to run and throw it in the lake because you can't think of any other safe place to put it. And then the water quality people start looking at you disapprovingly. So, especially since weight isn't an issue this year, I'm gonna stick with NiMH or similar for the rest of the year, and I urge other careless people to do the same!

Yeah, now I'm using a lead acid, incredibly heavy but has the highest current

[Ron_Swanson]

To answer your questions chalker:
1. We had 21 teams in our region and 15 competed in robot arm
2. Winning score was 94 (perfect score) using 6 motors. As I mentioned before, I would estimate that the top four scorers in our region were at least capable of getting a perfect score.
3. Mean score was 33.5 Median score was 19.
4. Four teams did not score any points. One was due to controller/power malfunction (as Jeff mentioned). Basically three minutes of pressing buttons and nothing happening. Two of teams simply did a bad job of anchoring the base and/or securing the arm to the base. On one of them the arm detached from the base almost as soon as it started moving. The other one the base left the robot square almost as soon as it started moving. The last one simply could not get itself out of the "ready to run position". It was clear that the way they designed it the motor which lifted the arm had just barely enough torque to do so and on that particular day it seemed that the batteries just didn't have enough juice.
5. I would say that the documentation was generally split 50/50. About half the teams had quality typewritten documents. They had clearly looked at the example documents on the SO website and followed the format closely. The other half had rather sloppy looking documents. A few of the teams were also missing a document. In general the quality of the documents mirrored the quality of the robots (i.e. the teams with the best robots also had the best documents).
6. I would say that the difficulty is adequate. The pencils seemed to be the hardest thing for the top teams. Probably because they are so close to the sideline to begin with. The number of objects seems about right in that you need a quick but also precise mechanism for moving the arm in order to do everything in time. Even the team that had a perfect score didn't have more than about 20 seconds remaining if I remember correctly.
7. The main change I would like to see is using time as a tiebreaker. I feel like motors will not break a lot of ties at the highest level (I foresee several perfect scores w/ 4 motors at nationals) and I don't like the documents tiebreaker since it seems too subjective. I agree with the idea of having the teams plan out the construction of their device and practice sound engineering principles. So I think that making teams submit documents (and having penalties for missing/incomplete documents) is totally reasonable. But I don't think its an appropriate tiebreaker for the top teams. This isn't a competition of who can make the prettiest CAD drawings. I think timing is a much better tiebreaker since it forces the best teams to address a critical engineering issue in their design (namely increasing speed while maintaining a level of precision necessary to pick up the pencils and PVCs without knocking them out of bounds) and it is a purely objective metric. Just like in sports how we hate to see a game decided by the referees, I would hate to see the national winner decided by the judges. Personally, I would like to see a system where the way ties are broken depends on the number of points scored:

For teams that get a perfect score:
time then motors then documents

For teams that get a nonzero nonperfect score:
motors then documents

For teams that get zero points:
documents only

To me that seems the most appropriate way to break ties at each level of robot ability.

[GoldenKnight1]

To answer your questions chalker:
.... Personally, I would like to see a system where the way ties are broken depends on the number of points scored:

For teams that get a perfect score:
time then motors then documents

For teams that get a nonzero nonperfect score:
motors then documents

For teams that get zero points:
documents only

To me that seems the most appropriate way to break ties at each level of robot ability.

I like the tie breakers better that Ron_Swanson suggest though I would add time somewhere for the nonzero nonperfect score robots because if a teams score 60 points (and they know that is the best they can do) and they can time at 2 minutes I feel that they should win some tie breaker over a team that used the full three minutes to score 60 points.

To answer Chalker's questions: 1. About 50% at the beginning of Jan and now more like 67%. 2. At three different competitions the max score was: 65, 72, 94. These are in the order that they happened and I have seen many teams get better as the season has gone on. 3. At the last competition the mean score for the teams that competed was 30.88. Median 23.5. 4. 1 out of 10 or 1 out of 15. This was either because the robot was poorly put together, the base moved outside of the box, or the team did not know how to control the arm well enough. 5. The average team had 2 completed documents and 1 incomplete. About 1/3 to 1/2 of the teams seemed to take these documents seriously but the did not have them or through them together. 6. The difficulty level is appropriate. 7. The pencils seem to be a problem for whether they went out of bounds or not. It would be better if the field was on a raised surface (instead of surrounded by tape) and if the objects touched the floor, they would be considered out. And again I like time as a time breaker at some level.

[wlsguy]

Hello Chalker7,
Since you are looking for suggestions....

It sounds like there are quite a few good robot arms and many not so good.
Robot arm is one of those events which requires significant expense to do well.
This tends to favor the larger schools and teams who have more money.
Solon, Mentor, and Chardon (the schools who had some of the best scores at CWRU) all fit this category.
The "lesser" schools, obviously couldn't compete. I would be interested why not all teams participated.
Is this because of money or something else; that's obviously a discussion point.

Now to give credit, some building events don't have a huge cost and can be competitive on $50 or less
examples; Sumo, robot ramble, gravity vehicle, mousetrap, trajectory, Wright Stuff, etc

Please consider those schools whose TOTAL Science Olympiad budget for events is ~$500 / year so all teams can participate.
I think this should be the target for new investment each year (i.e. supplies for the new events).
All schools should be able to compete at a similar level with a reasonable amount of money.

Science Olympaid should be more about effort and less about $$$$.

[chalker7]

Hello Chalker7,
Since you are looking for suggestions....

It sounds like there are quite a few good robot arms and many not so good.
Robot arm is one of those events which requires significant expense to do well.
This tends to favor the larger schools and teams who have more money.
Solon, Mentor, and Chardon (the schools who had some of the best scores at CWRU) all fit this category.
The "lesser" schools, obviously couldn't compete. I would be interested why not all teams participated.
Is this because of money or something else; that's obviously a discussion point.

Now to give credit, some building events don't have a huge cost and can be competitive on $50 or less
examples; Sumo, robot ramble, gravity vehicle, mousetrap, trajectory, Wright Stuff, etc

Please consider those schools whose TOTAL Science Olympiad budget for events is ~$500 / year so all teams can participate.
I think this should be the target for new investment each year (i.e. supplies for the new events).
All schools should be able to compete at a similar level with a reasonable amount of money.

Science Olympaid should be more about effort and less about $$$$.

We hear this argument every year and are extremely sensitive to it, the main question is how to limit costs while keeping the event exciting (which, I assure you, we are trying to do). The only thing I am curious about, how is Robot Arm more expensive than Sumo? To me, that event was much more a race to see who could buy the nicest parts than Robot Arm is.

[ohiostar]

Being from one of the larger schools mentioned above, I'd like to add that there wasn't a great deal of interest in this event early on, we only had 3 students sign up. We ended up being competitive with a robot built from disassembled Sumo Bots. The team was very frugal and obtained materials mostly from family garages and basements, re-purposing many items. A second robot arm as built with hobby servos and moderately priced components available at many online sources.

Contrary to what many believe there is not a crazy amount of money available for building events. Most of our funds are used up for traveling to invitationals etc.

[ODoyleRules]

Having ran Sumo, Robot Ramble, Robot Arm, and just about every other Science Olympiad Robotics event at Invitationals and Regionals over the years, I can say that this really does seem like the most expensive Robotics event I have ever seen at S.O... I think one of the big differences for many teams is that Sumobot could be done reasonablely well with VEX kits. I mean even at Nats last year, there were clearly several robots there that were built on VEX frames, and many of them did fairly well. VEX kits do not lend themselves to Robot Arm as well. Most the VEX robots I have seen this year were both slow, and incapable of picking up the D battery and putting them in a goal box. Teams I think are starting to realize this, because I am starting to see fewer and fewer VEX bots at competitions. This means purchasing Survos and related parts to run their Robots, and more cost to the team.

In my book, the most reasonable S.O. robotics event we have ever had was Robot Ramble. It could truly be done on a shoe string budget. I know the State Champions in Indiana, the last time Robot Ramble was an event, had a perfect run in 27 sec, and the total cost of the robot was less the $50...

[chalker]

Just wanted to chime in on the cost issue. As my brother indicated we are very sensitive to this and have looked at it closely. I think it can be done for a quite reasonable cost yet still be competitive if you know where to look. For example, I've spec'd out what I think would be a reasonably competitive 6DOF robot (gripper, wrist rotate, wrist flex, elbow, shoulder, and waist) with the following (all new, not used, parts):
-A 6 Channel RC transmitter / receiver for $23.
-Servos that work with it are $2.60 a piece and have 2.5kg of torque ( which should be more than enough for a basic robot).
-Add in batteries for ~$4
-Maybe another $10-$15 for wood / pvc / metal strips to build the structure of the robot

That gives you all the parts you need for ~$55 (note some assembly required;).

I think the problem we are finding is that too many teams want to go buy kits or things like Mindstorms / Vex and just 'snap together' a design, which will inherently be more expensive since someone else did all the engineering work already for you. We don't have that same problem in most of the other building events, so I'm not sure why it is an issue here. Any ideas on how we can change the rules to push teams more towards the building from various parts?

[ODoyleRules]

Just wanted to chime in on the cost issue. As my brother indicated we are very sensitive to this and have looked at it closely. I think it can be done for a quite reasonable cost yet still be competitive if you know where to look. For example, I've spec'd out what I think would be a reasonably competitive 6DOF robot (gripper, wrist rotate, wrist flex, elbow, shoulder, and waist) with the following (all new, not used, parts):
-A 6 Channel RC transmitter / receiver for $23.
-Servos that work with it are $2.60 a piece and have 2.5kg of torque ( which should be more than enough for a basic robot).
-Add in batteries for ~$4
-Maybe another $10-$15 for wood / pvc / metal strips to build the structure of the robot

That gives you all the parts you need for ~$55 (note some assembly required;).

I think the problem we are finding is that too many teams want to go buy kits or things like Mindstorms / Vex and just 'snap together' a design, which will inherently be more expensive since someone else did all the engineering work already for you. We don't have that same problem in most of the other building events, so I'm not sure why it is an issue here. Any ideas on how we can change the rules to push teams more towards the building from various parts?

I can put a bunch of prices up too for parts, but that still does not change the reality of what teams are spending. Also, your list is unrealistic. First of all a servo with only 2.5 kg of torque will not have enough torque at the base to put a battery in the north goal. Especially if it is built by your average student. I have ran this event at 4 different invites this season, and will be running it for my fifth tomorrow, and I can tell you. No one has won any invtational I have been at for less then $500. I ask every team that comes in how much they spent on their robot, and in general they are all well over $100.

[chalker7]

I can put a bunch of prices up too for parts, but that still does not change the reality of what teams are spending. Also, your list is unrealistic. First of all a servo with only 2.5 kg of torque will not have enough torque at the base to put a battery in the north goal. Especially if it is built by your average student. I have ran this event at 4 different invites this season, and will be running it for my fifth tomorrow, and I can tell you. No one has won any invtational I have been at for less then $500. I ask every team that comes in how much they spent on their robot, and in general they are all well over $100.

We definitely know the reality of this event is that teams are spending a large amount of money, but the larger point is that they are not required to do so. I built a functioning robot arm using components below the specs quoted by my brother that was strong enough to lift D-Cell batteries, it just required a fair amount of design work and engineering planning (you don't need a servo at the hub lifting the whole arm if you use a crane or SCARA style arm.)

Our larger point is that in the other building events (with the possible exception of helicopters), kits or predesigned tools are relatively rare. Why is it that teams always revert to such components in the robotics events when far cheaper components are available? I have a few theories, but I won't presume anything in this situation.