pic: Cyber Blue Swerve Chassis - Rear View
 

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Looks extremely slick, kudos to 221 LLC and Cyber Blue. What kind of weight are you guys looking at for that setup right there?

Brando

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

To be honest, we have no idea how much it's going to weigh. Our focus so far has just been "get it done" so that the programming team can start working. Once it's finished, which it is now, we will look at ways to save weight for potential future usage, as right now it is way too heavy (estimation) to use as a competition robot

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Yeah I had no intention of starting the "you can save weight here" discussion, I just wanted to know if you had a rough idea of that system. I know the one that was posted on 221's site was 100+ lbs. but that was for a client (non FRC). It's always nice to compare weights of different configurations to see what could be more beneficial. But seriously, it looks really nice thus far.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

What sort of sensors are you using for the angle measurement?
EDIT: Never mind, they're clearly potentiometers or something that acts as such (3 - pin)

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

<strike>Potentiometers, as can be seen here</strike>

Out of curiosity, if you are going to connect all 4 modules together why bother with 2 Globes?

[insert question by Dave about how well you can adapt legs onto it]

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Redundancy. I seem to recall that at one point in 2005, a PWM for an arm motor's controller somehow disconnected on 330's robot. We didn't notice for I don't know how long, simply because the other motor on the arm was enough. In 2007, with the same design, we removed one arm motor to make weight with both ramps.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

As good a reason as any I suppose. Thanks.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

[quote=Andrew Schreiber;880335
Out of curiosity, if you are going to connect all 4 modules together why bother with 2 Globes? [/QUOTE]

Yes, for redundancy and accuracy.

Redundancy - One globe has the power to turn all for modules, two will make sure they will turn even if one fails or loses power (connector, software, etc.)

Accuracy - By tying all four together, we are assured that each wheel turns the same number of degrees all at the same time. Keeping the left side and right side separated would allow for error in the systems to let one set of modules turn a little more or a little less and possible not return to "0" after a turn.

Also, this is a prototype / demo and some of the precautions and redundancies we are putting in might not make sense on a competition robot where you are making the risk / complexity / weight trade-offs.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Looks great.

One concern - I cannot tell from the picture how the globe motor is mounted or how the sprockets are mounted to it. I've always understood that globe motors are not designed to have a cantilevered load on the shaft coming out of its gearbox.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Raul -

That is a legitamate concern. The globes are mounted to an aluminum plate which is mounted on the bottom of the top 80/20 rail. We then use the Globe DD hub from AndyMark (here) with two sprockets and spacers.

I, personally, don't have a ton of experience with the Globe motor, but the team does. If this is true of the globe, it wouldn't be too difficult to build a support structure for the top of the Globe/sprocket assembly.

How does 111 turn their crab modules?

Thanks for your thoughts. We are building this now so that we may learn from those with more experience than us without the stress of a build season. Any other questions or criticisms are welcome.

//Edit//

This photo may show the Globe mounting a little better.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

It looks extremely clean and uncomplicated. Definitely not ours. (we rushed ours into service four weeks into build after NO previous experience with crab/swerve drive xD) Keep the good work going. I'm looking forward to seeing more pictures in the future.

P.S. I have access to our drive frame if you need it.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

I agree with Raul, and always support the output shaft of the Globe motor.
Maybe I'm overly cautious? Let me know how your prototype works unsupported.

-John

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

I can see why to be cautious, yet even that being said whenever my team has used globes the other end has been unsupported and hasnt given us any issues.
But the places in which we used them didnt have constant sideloads, just occasional ones.
In theory the chain going both ways off of it may cancel out some

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Looks real nice.

Here's a tip that will save you time and money. If those pots have hard stops at each end and can be broken by being driven past those stops, add some mechanical stops to the system. Your electrical team will thank you. The system will get away from you at least once during early software development and will start spinning wildly. It's just something I've heard and not from any personal experiences :rolleyes:

Can you share with everyone how you plan on using the data from both pots? We've contemplated adding redundancy in the past, but have never come up with an easy way to distinguish a good vs a bad pot so we always stick with one.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

We talked about this last night and are trying to determine where is the best place for hard stops. We did not find a place yet, but your comment reinforces the desire to have something in place.

I will talk to our programming leaders and get a response. I don't know if they plan on averaging the input data or using one as a primary and one as a backup or something else totally different.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Love the work. Are you going to have a demo time?

My question about the two globes and the redundancy: If one globe dies will the other have enough power two turn the modules and overcome the drag from the other globe?

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Don't know about a demo - maybe something at the IUPUI kick-off?

As long as globe1 doesn' fail and actually lock-up, then yes, one will overcome the other. Last night we put power to just one motor at a time and it (quickly) rotated the modules. We expect we will have to really limit the power to the globes to be able to control the rotation.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Collin,
Same here for us. We started during the summer, our swerve project which is an off-shoot of what 1983 Skunkworks created.
Our thoughts were to go through the grind now rather than during build season, in case we use it or not. At least you got options come Kickoff.

With so many teams developing something last season, the last thing we wanted to do was be 2 years behind, in case it would be beneficial to have such a drive system for the 2010 game.

Nice work and thanks for the better photo.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

These units make the steering job easier, Wild Swerve Steering Module. Though they don't support two sprockets.

In my discussions with Raul he has mentioned that they commonly run a 2.5:1 reduction from the globe to the module.

The chassis looks great by the way. We've thoroughly enjoyed working with Cyber Blue. Their work on the assembly instructions for this unit is very impressive; as is their dedication to educating the FIRST community.

Thank you.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

How is the chain tensioned?

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

threaded standoff setups with the chain masterlinked to them, only works for these cause the modules cant go round and round like coaxial can.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Yes. The tensioners are basicaly turnbuckles, with a drilled and flattened end on the screw threads to accept one side of a master link.

We made a dense foam mounting board with four wheel sized slots cut into it so that we could set the chassis onto it for alignment. After the wheels were aligned, we set the chain tension (the tops of the modules have some adjustment) and then used the tensioners on the other side of the sprockets.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

I actually just weighed our robot with another member last week due to curiosity. It weighed near 80 pounds. But that also includes our electronics. However, I don't recall whether or not the battery was on or off the robot at that time. The 80/20 chassis we put together was 18 and each pod is presumably 8 or 10 pounds each. So what you are looking at in the picture is about 60, give or take a few. Of course you can always make it a lot lighter in a number of ways.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Looks great, we had pretty much the same design for our competition bot last year. what's your sensor and driver station setup?

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

are both your front wheels and back wheels steerable?

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Yes. If you follow the chain runs, the sides (where the turning motors are) have a run to the far end and a run to the near end. Then there is a run across the near end for redundancy. All four modules rotate together.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Thank you.

Currently, the only sensors are the potentiometers on the rear wheels which monitor angular position of the pods relative to the frame.

These units from Team 221 LLC do come with an easy way to mount an encoder on each wheel, should you choose to do that.

We are using a PS2-style game pad to drive with one "joystick" controlling the angular position of the wheels (powered by the globe motors) and the other controlling the speed of the wheels (powered by the CIMs)

Overall the design has worked very well for us, and we're thankful that Anthony and Team 221 LLC decided to partner with us for this project.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

when we built our crab drive, one problem we had was that the motor would over shoot the intended position, then try to swing back, and the process repeats.

we multiplied the difference between the wheels position and the intended position (the value read from the controller, translated to the same value range as the wheel position) by a constant less than 1 (ours ended up being around .57, but this is something you can define at the top of your code and mess around with) this way your wheels slow down as they approach the intended destination.

how did you solve the problem? maybe we could trade some code

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

I am not a programmer, but I will tell you how I think we solved the problem. I believe we measure the delta between the angle we want and the current angle of the wheel. While that angle is >15 degrees, the wheel rotates at full speed. Between 15 and 5 degrees, the wheel rotates at a lesser speed. Between 5 and 0 degrees, it rotates slower, still.

We didn't seem to have any problem with overshoot. The wheels seemed to do a good job of "snapping" to a position and holding there.

All that said, I will check with the Controls team Monday and confirm this.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

This is a classic controls problem. You guys have implemented a "P" controller. Adding a derivative term will help, and make it a "PD" controller. One way to tune the system is to increase Kp until it gets pretty close to oscillating constantly, and then increase Kd until it doesn't overshoot any more. If at any point it starts acting "funny", decrease the constant some. One thing that really helps is to get back the position data at each point in time and then plot it.

I believe that there are some white papers around, but until I finally got the chance to take a controls class and talk with someone who does controls for a living, I had a little trouble with tuning the loops too.

I've used the following successfully many times in the past.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

well its not so much an overshooting that we had problems with, it was that the motors would start vibrating because they were trying to switch directions so fast. it seemed harmless at first until the motors started getting hot. if you multiply the delta angle by a constant (where constant < 1) like i was describing it should give you a little bit better response

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

Actually, to correct Collin a little. we are currently using the P term of the PID control with very little or no overshoot. However, we do have a maximum and minimum speed and a small deadband around the target angle. This seems to work pretty well on the bench but we haven't tried it on the floor yet. I know it will be different on the floor but I'm wondering if we will see a difference when the robot is standing still and when it is moving.

Re: pic: Cyber Blue Swerve Chassis - Rear View
 

we got an official weight tonight. The robot with no battery and no bumpers, weighs 77 pounds.

I am not sure if this was already posted here, but we also have a video up on youtube of our first practice session with the robot.

that can be seen here

Matt G.