2012 FRC Team 1717 Uncut

[jakemochas]

Originally Posted by Siri

Can I ask what drive & steering motors did you use in 2009-2012?

Below I have provided an overview detailing our swerve drives for the years 2009-2012:

2009
This was the first year that we decided to use a swerve drivetrain. Our drivetrain was split down the middle of the robot. On the left side, both of the drive wheels were paired together and powered by 1 CIM motor through a single speed belt reduction that used timing pulleys. The left wheels were also paired together for turning. The turning was powered by a banebots RS-545 though a banebots planetary transmission.

The right side of the robot was a mirror image of the left side.

2010
The drive for the wheels were again coupled together in pairs. Both of the left wheels were driven together and both of the right wheels were driven together. To power this drive, each pair of wheels was powered by two CIM motors through a two-speed transmission that was a re-packaged AM supershifter gear-set with some custom modifications.

For turning, the front left and back right wheels were paired together and the front right and back left wheels were paired together. This enabled both translational motion in any direction and rotation of the robot about its origin when it was not translating. Each turning pair was powered by a single fisher price motor through our custom gearbox with purchased gears.

2011
The wheel pairings were the same as in 2010, but the turning motors were banebots RS-775 motors.

Our drive transmissions were custom two-speed transmissions that were optimized for our swerve system. The turning transmissions were custom as well. Also, we cut all of our gears in house and they were made from steel.

2012
In 2012, our drivetrain consisted of four independent wheel modules. Each wheel module’s drive was powered by one CIM motor with a two-speed custom gearbox. Our wheel module’s turning was powered by a single banebot RS-550 with a custom transmission. All of our gears were made from aluminum and they were cut in house.

In an upcoming post we will provide some pictures of the 2012 modules and describe them a little more thoroughly.

[Steven Donow]

Originally Posted by jakemochas

In an upcoming post we will provide some pictures of the 2012 modules and describe them a little more thoroughly.

And the countdown for what could be one of the greatest posts ever begins...

[Akash Rastogi]

Jake, what method were you using to cut your gears in house? Are they stacked plate gears?

Also, you said your team had 2 speed transmissions optimized for your swerve, what iterations did your transmissions go through in order to optimize for a swerve?

Thanks for all the great insight!

-Akash

[R.C.]

Originally Posted by Akash Rastogi

Jake, what method were you using to cut your gears in house? Are they stacked plate gears?

I believe it was cut using the 4th axis (Tormach CNC?). They were definitely not stacked gears and were pocketed very nicely.

-RC

[jakemochas]

Originally Posted by Akash Rastogi

Jake, what method were you using to cut your gears in house? Are they stacked plate gears?

Also, you said your team had 2 speed transmissions optimized for your swerve, what iterations did your transmissions go through in order to optimize for a swerve?

Thanks for all the great insight!

-Akash

Originally Posted by R.C.

I believe it was cut using the 4th axis (Tormach CNC?). They were definitely not stacked gears and were pocketed very nicely.

-RC

We cut our gears using a Tormaq CNC mill equipped with a 4th axis. The gears are cut using involute spur-gear cutters. Below is a McMaster-Carr link where you can order the type of gear cutters we use. I have also included a picture of a gear cutter below.

http://www.mcmaster.com/#gear-cutters/=i19i2u



We buy round stock and from that we make gear stock. I have included a sample picture of a piece of gear stock below.


In order to make the gear stock, we use the process shown in the video link posted below. In the video, however, they are making a single gear. We use this same process to make an entire piece of gear stock. After we make the gear stock, we part it off in a lathe to make the individual gears.

http://www.youtube.com/watch?v=sCuf6RqM7e0

[JamesTerm]

Originally Posted by jakemochas

We did not use a magnetometer. We only used a gyro.

In the end, though, we were not satisfied with any of the gyros. The gyros had significant drift, asymmetrical behavior, and were extremely sensitive to collisions on the field. In about 45 seconds, the gyro would drift by as much as 60 degrees and the advantages of the gyro would be lost.

In order to solve this problem....

One of our goals for next year is to find a better gyro solution for our swerve drive system.

Ok I'd like to throw an idea to you, and others feel free to shoot holes through this (couEthergh cough cough).


I'm going to start out with two assumptions that I hope people can help me verify:

1. A gyro... has quick repsonse, but is subject to drift
2. A Magnetometer does not drift but has a slow response (lag)

If these assumptions are correct (to the way I'm thinking they work), then I'd propose the idea of having them work together to yield a quick reponse heading that remains fairly accurate.

I've solved this kind of problem before at work and since you are a c++ programmer... I'll show you the code where I do this. right here:

 			size_t GetPlayPos()
			{
				DWORD playpos;
				//size_t SampleOffset=0;
				size_t SampleOffset=(size_t)(0.038 * m_SampleRate);
				m_lpdsb->GetCurrentPosition(&playpos,NULL);
				playpos/=m_BlockAlign; //convert to samples

				//Note: This design assumes the buffer size (of our secondary buffer) is the same size as the sample rate.  This is fine for now, but at some
				//point I may need to make a distinction if I have to change the buffer size.
				
				time_type CPUClock=((__int64)time_type::get_current_time()+m_ClockPhaseOffset) % 10000000;
				double CalibratePlayPos=(double)CPUClock*m_SampleRate;
				int PhaseOffset=(int)playpos-(int)CalibratePlayPos;
				int iSampleRate=(int)m_SampleRate;
				//Check wrap-around case
				if (PhaseOffset > iSampleRate>>1)
					PhaseOffset= ((int)playpos-iSampleRate) - (int)CalibratePlayPos;
				else if (PhaseOffset < -(iSampleRate>>1))
					PhaseOffset= playpos - ((int)CalibratePlayPos-iSampleRate);

				double Current_ClockPhaseOffset=(double)m_ClockPhaseOffset + (((double)PhaseOffset / m_SampleRate) * 10000000.0);

				//check math
				//CPUClock=((__int64)time_type::get_current_time() + (__int64)Current_ClockPhaseOffset) % 10000000;
				//CalibratePlayPos=(double)CPUClock*m_SampleRate;
				//printf("Test-> playpos %d, %d \n",playpos,(size_t)(CalibratePlayPos));

				const double dSmoothingValue=0.1;
				//blend the current phase error to the current phase offset
				m_ClockPhaseOffset=(__int64) (((1.0-dSmoothingValue) * (double)m_ClockPhaseOffset) + (dSmoothingValue * Current_ClockPhaseOffset));
				//debug_output(p_debug_category,L"playpos %d, %d, %d\n",playpos,(size_t)(CalibratePlayPos),PhaseOffset);
				size_t Error=abs(PhaseOffset);
				//printf("\r%d          ",Error);
				//if (Error>1000)
				//	debug_output(p_debug_category,L"%d\n",Error);

				return AdvancePosition((size_t)(CalibratePlayPos),SampleOffset);
			}

In this example we want to determine the play position of the audio cursor inside a sound card (using direct sound). The reality is the GetCurrentPosition() method is similar to using (what my assumption is of) the magnetometer where it has lag (about 20 30ms of repeated results). This is not acceptable! So to solve the problem I use the PC clock since it has a quick response (e.g. like the gyro). The way this code works is it applies a continuous error correction to whatever the current clocktime is read... so it would be like using the gyro reading as your base answer and applying the magentometer error correction on it. The design here is similar to using P in PID (in the PhaseOffset variable)... and doing a blend function on the error computed. The blend allows the gyro to have more influence on quick movement and less influence on slower movements while over time it is always recalibrating itself to the correct point of reference. I hope this idea is useful and the assumptions about the sensors are correct... I may have a run with these.

[rachelholladay]

(Sorry to totally deviate from the robot design discuss, but i have a question..)

Originally Posted by jakemochas

As you may know, our team is made up of all seniors and we only get to do FIRST once.

Maybe because this is so radically different then the way my team is structured, but this statement seemed a little odd to me. Is it by design or accident that the team is all seniors? If by design, then why? I am very curious..

[msimon785]

Originally Posted by rachelholladay

(Sorry to totally deviate from the robot design discuss, but i have a question..)



Maybe because this is so radically different then the way my team is structured, but this statement seemed a little odd to me. Is it by design or accident that the team is all seniors? If by design, then why? I am very curious..

It is by design. On 1717, 9th, 10th and 11th grade are spent in preparation for the culminative senior year that is FIRST. I know in 10th and 11th grade, you have the option of coming to competitions (I don't remember the details, though), but your only experience building an FRC robot is in 12th grade.

IIRC, each grade focuses on a specific aspect of engineering - so the students get a well-rounded background - before they participate in FRC.

[jakemochas]

Hi everyone,
I am trying to answer all of your questions and I have many responses ready to go. At the moment, however, the thread will not let me post links or images to the forum. Many of the answers would be incomplete without this content. We are working as quickly as possible to get your answers up!
Thanks,
Jake
FRC Team 1717

[JamesD]

Originally Posted by msimon785

It is by design. On 1717, 9th, 10th and 11th grade are spent in preparation for the culminative senior year that is FIRST. I know in 10th and 11th grade, you have the option of coming to competitions (I don't remember the details, though), but your only experience building an FRC robot is in 12th grade.

IIRC, each grade focuses on a specific aspect of engineering - so the students get a well-rounded background - before they participate in FRC.

As I've heard, the FRC competition in the senior year is the culmination of the 4 year program. They aren't just a robotics team, but an engineering academy.

The BeachBots have had the pleasure of knowing and competing with team 1717 for several years now and have seen them grow into a powerhouse team. They continue to improve each year and are wonderful competitors and friends.

On May 26th the D’Penguineers had an open house at their new engineering academy building in Goleta and invited the BeachBot team, as well as many others, up to visit. Many of us took the drive (3.5 hours that should have been 2) to go visit with them.

As many of us know, Amir's program and foundation at Dos Pueblos High with which he created the "Dos Pueblos Engineering Academy" is a great example of what can be done to promote, inspire, and educate students, not just in engineering, but in many disciplines that have a significant contribution to our culture and future.

I don't know if it's all in the “New Cool” book about them or not (that’s still on my reading list), but if you have any interest in replicating what they are doing (and you should) I think these notes will give you a bit of an idea of how to go about it.

================================================== ==========================

Engineering Academy notes:

Key: Curriculum is approved by the University of California schools so it counts toward entrance requirements. Parents prefer their kids take courses that help them get into college. Without this approval it's a nice program, but parents won't invest as much into it and won't want their kids to invest time in it since it wouldn't count towards college entrance requirements. [There are many universities that recognize FIRST students and offer scholarships to them, but the time invested with the team is not often recognized as satisfying a university entrance requirement]

Steps to make it happen (how it can be replicated):

1. It starts with someone passionate about it to drive it (thanks Amir!).

2. Build a board of directors - a few or all of these people are to emphasize funding and organization. They provide the Board of Directors for the 501(c)3 corporation.

3. Create a 501(c)3 organization - the foundation. This includes creating a mission statement and creating the financial structure needed (bank accounts, double signature approval process), financial
accountability, etc. A lawyer and an accountant on the board would help greatly with this.

4. Study and replicate DPEA's curriculum: Art, Science, Math, Engineering (technology). Art is important as it leads to design and creativity. DPEA (Amir) has done all the legwork, just replicate his curriculum, he seems glad to share it. Again, the curriculum is critical as without the UC approval it's just a nice club. Note that FIRST robotics is a part of the program for the second half of the senior year. It’s not a robotics program but an engineering program. FRC is just the last project (in a long line of increasingly more complicated projects) for the seniors.

http://www.dpengineering.org/academy/plan/four_year2

- Read the example projects and course descriptions on this page (really, take some time to review this stuff):
http://www.dpengineering.org/academy/plan/courses2

5. Design a plan: building needs, number of kids, equipment needs, personnel needs - paid and qualified teachers and engineers. Finding or training the right teachers will be important. Even to the point of having them spend time with Amir if possible. Define a budget, organization, and implementation schedule.

6. Begin to seek funding, both initial and on-going commitments. Develop a presentation for fundraising that includes detailed goals, costs, budget, etc. Include some students in the presentation to local businesses, corporations and foundations. 10 minutes long, no more. In person presentations if at all possible.

7. Develop a website and marketing materials leading into step 8 below.

8. Build energy in the community. Promotions, awareness, value to kids and community. I think this too is critical. DPEA works well since Santa Barbara is a small town community (well to some of us). Marketing to the local businesses, television station, newspapers, and the people in general works better since they all identify with the community (businesses want to be seen as participants, giving back; people want to see local schools, teams, doing well; etc.) In a metropolitan area you would have to define what your support community is and build lines of communication, marketing, awareness, value, and excitement about the project, etc.

9. Based on funding, create a timeline and identify resources needed at each point. Then for sustainment.

10. Begin implementation. Identify qualified instructors. Retired engineers? Possibly even seek donation of current engineer's time from corporations (4 hours, twice a week, who knows). Art teachers. Math & Science teachers.


Advantages from using Amir's work:
- curriculum approved by UC schools so it counts toward entrance requirements
- he's already done a lot of the leg work on all this
- visibility of his program

I think it would be good to have those at DPEA take a look first, to see if there's anything I missed that they think would be helpful to add. Could be something big I'm overlooking that they could point out.

[rachelholladay]

Originally Posted by msimon785

It is by design. On 1717, 9th, 10th and 11th grade are spent in preparation for the culminative senior year that is FIRST. I know in 10th and 11th grade, you have the option of coming to competitions (I don't remember the details, though), but your only experience building an FRC robot is in 12th grade.

Why? (Personally, it would break my heart to only do FRC one year)

[Steven Donow]

Originally Posted by rachelholladay

Why? (Personally, it would break my heart to only do FRC one year)

When reading The New Cool, it's easier to understand why. These kids are basically trained for FRC. Their senior year, school is FRC. And it's not like prior to senior year they're just learning engineering, from what it seems like, they basically are doing FRC-like things.

A thought I've always had as to "Why 1717 haven't been to Einstein/been champions" is the fact that they're all FRC rookies every year. But they're not. They spend 4 years completely surrounded by FRC.

Plus, as the poster from 1515 said, they(1717 students) do go on to mentor.

[dcarr]

I had a few questions regarding encoders:

We're working on our drivetrain over the summer. The past two years we have used CAN with our Jaguars, connected to a 2Can, but it has been riddled with problems. So we are switching to PWM which I know you guys use and plugging the encoders into the sidecar. Since each encoder can take 2 I/O ports, how did you manage that-did you fill one up? (if so, did you use more than one sidecar?) How did your sensor arrangement work, did you use many others besides the encoders and the gyro?

Thanks for taking the time to answer these questions, it is greatly appreciated!

[dcarr]

One more thing...the swerve modules seem to rotate a lot as you stop and prepare to shoot. Is this part of re-zeroing the gyro for each shot, or some other reason?

[jakemochas]

Hi everyone! Our answer to the gear cutting question has been approved and can be seen above at post #65. The post is above because it was posted on the 18th, but it didn’t show up until today. We are working hard to post all of our other answers as well! Thank you for your patience.

Originally Posted by dcarr

I had a few questions regarding encoders:

We're working on our drivetrain over the summer. The past two years we have used CAN with our Jaguars, connected to a 2Can, but it has been riddled with problems. So we are switching to PWM which I know you guys use and plugging the encoders into the sidecar. Since each encoder can take 2 I/O ports, how did you manage that-did you fill one up? (if so, did you use more than one sidecar?) How did your sensor arrangement work, did you use many others besides the encoders and the gyro?

Thanks for taking the time to answer these questions, it is greatly appreciated!

In order to PWM our 14 motor controllers and have enough I/O for the encoders, we did use two digital sidecars. As a result of the new motor allowance this year, it was the first time we used two digital sidecars.

We had many sensors on our robot that were wired to the analog and digital slots of the cRIO. For our swerve drive, we used two types of sensors: we used magnetic encoders on the analog board for wheel orientation and optical encoders for the wheel drive. Also, the gyro was attached to an analog port.

For the shooter, we used a magnetic encoder attached to the analog breakout to detect the platform position. For the hood angle, we used a multi-turn potentiometer on the analog breakout. To detect flywheel speed, we read from a light sensor that was attached to the digital sidecar I/O.

Originally Posted by dcarr

One more thing...the swerve modules seem to rotate a lot as you stop and prepare to shoot. Is this part of re-zeroing the gyro for each shot, or some other reason?

The rotation of the wheels when we prepare to shoot is mostly from re-zeroing, but some of the turning is from fine adjustments the driver makes while lining up for the shot.