Effectiveness of Casters

[Spikey]

Originally Posted by Gary Dillard

Spikey, I'm impressed that you won 2 regionals with casters, but I'm betting it's not because of the casters right? You just focused your resources on other parts of the robot rather than on the drive system, so you had a robot that played all aspects of the game and didn't just drive.

The question of this thread is, what do you think of casters. My answer - not much.

Our robot that year was very fast and maneuverable, we could fly across to the other side of the playing field and steal our opponents black ball, then quickly collect our own giving us a huge advantage. Also we had a very effective scissor jack to deposit balls, and to hang on the bar. Casters actually ended up as a last resort for our team because our 4 wheel drive destroyed the carpet, after we slapped the casters on we went from being dead last to winning two regionals. However this year will not be a good year for casters because they do not do well at all on the polycarbonate(u need traction on there) and the difficulty they add when you have to climb stairs(getting caught on them). This year is a bad year for casters.

[Rickertsen2]

Originally Posted by KenWittlief

where are all the REAL engineers in this thread? doenst anyone use feedback to control their robot?

yes when you put castors on a bot the steering gets squirrely - thats why you also use a yaw rate sensor that measures how much the bot is turning, and you fly-by-wire!

you close the loop on steering by having the SW look at what the driver is commanding the bot to do (how hard driver wants to turn) and looking at the yaw rate sensor to see how fast the bot actually IS turning

and use a PID closed loop algorythm to MAKE the bot do what its told!

why would you do anything else? this works with 2 wheel skid steering, 2 wheel castor steering - four wheel steering - 6 wheel.... whatever the gearheads throw at you, the sparkies can turn the machine into a PID controlled nice and tight, highly responsive servo like machine

the beauty of this is not only does it make the robot go straight when you want it to go straight, it also allows very precise slow turns. If you want the bot to turn just a tiny bit, the SW will put the necessary power to each motor to make that happen, adjusting the levels 40 times a second.

Try it on ANY bot - feedback is the most powerful tool an engineer has at their disposal - see them F111s flying around, they would be completely unstable, unable to fly AT ALL without fly-by-wire closed loop feedback controlling its motion. And guess what? the same type of yaw rate sensors they use are available to you in the suppliers catalogs.

Hah. Our robot already has 3 PI controllers and we Havn't even started codign for THE MANIPULATOR yet. Feedback control rules!!

We still wouldn't touch castors again after our last experience with them.

Please!! If you are considering casters, at least use skid plates instead.

[ngreen]

Originally Posted by Cory

The only reason that I would ever stick a castor on a robot would be to help turn. I would use a cylinder to engage or retract it, but I would never let two of them be in constant contact with the ground. If you had been around in past years, you would see robots with two driven wheels and two castors like you said. Basically, they would spin around like crazy, with very little control.

Cory

Yeah, castors that extend and retract are great for a robot with four wheel and too much traction. We were tripping breakers when we turned until we put the casters on. Just a tip, get sturdy casters and if you use pneumatic cyclinders do everything you can to protect the rod. They don't like side loads and the machine shop guys get tired of bending them back. Also, only lift so that the turning wheels touch and the others barely touch.

[Aignam]

Team 103, the Cybersonics, used two casters in their 2003 robot. That machine gave more or less every robot a run for their money. I'm not too certain of the specifics, or how it worked, but you most certainly can build a competetive and successful robot using casters.

[Pat Roche]

Our team tested a prototype drive train/robot at River Rage and I drove all the matches and I can say from experience you will have trouble controlling it with two wheel drive and casters. Remember though that casters arent bad when used in combonation with tracks/tread or 4/6 wheel drive. We competed well with a caster to support or robot with tracks last season.

Also remember to acount for what you actually using them for. They do add weight to the bot.

Just some thoughts to dwell upon,

-Pat

[ShadowKnight]

other people have brought up valid concerns that could lead to certain teams deciding to not utilize casters in this year's game. As far as a robot that intends to climb the stairs, casters WILL become a hangup and possibly could cost the team points and time in the pits try to fix their messed up bot because the casters broke off. Casters also are bad bad bad for driving if used as 2. 1 is MUCH better because of how the drive train pivots. so...I personally see casters as a design flaw that will leave a team hanging on a hangup. Having driven bots with 2 casters, I can tell you that they are crazy bad on anything but carpet and are not that easy to control with precision...