Drive systems

[Roboticsismylif]

Hey everyone. This is our second year as a team. Right now our team is split on what we should use for a drive system. Most of the team, including me, want to go with tank treads. The others what to go with wheels. Right now we're going with a defensive robot. Anybody got any hints and suggestions?
Thanks.

[pfreivald]

Originally Posted by Roboticsismylif

Hey everyone. This is our second year as a team. Right now our team is split on what we should use for a drive system. Most of the team, including me, want to go with tank treads. The others what to go with wheels. Right now we're going with a defensive robot. Anybody got any hints and suggestions?
Thanks.

If you haven't prototyped tank treads pre-season, I wouldn't recommend them now. Often times newer teams do not anticipate how much torque is required to turn when working with high-traction systems.

Originally Posted by Roboticsismylif

Hey everyone. This is our second year as a team. Right now our team is split on what we should use for a drive system. Most of the team, including me, want to go with tank treads. The others what to go with wheels. Right now we're going with a defensive robot. Anybody got any hints and suggestions?
Thanks.

Listen. Everything I'm going to say is in your best interest, with the sole intent of helping you. I want you to understand that.

DO NOT BUILD A DEFENSIVE ROBOT. I don't care what your team says, show them this. KatieUPS will agree with me, ask her. If you build an offensive robot that ends up not working, you can always play defense. If you make a defensive robot that ends up not working, you're screwed. You want to be able to score in some way possible each year, no matter what.

For starters, if you DO do defense this year, you better have a good as (excuse my language) hell defensive strategy. If you are on the opponent's side without knowing what you're doing and being able to execute it 100% of the time, then you'll be getting more penalties than you're worth. Defense this year will be extremely difficult to do unless you are prepared for it. I'm not saying you aren't, but even for prepared teams it'll be tough and risky.

My solution is this: Create a way you can at least gather balls and shoot/throw them. You may be able to only score on the lowest hoop in autonomous and feed your team balls the rest of the match, but it's a lot better than defense this year, and a lot more valuable.

If you have any questions, please feel free to PM/contact me for ideas/designs. We're chalk full of them, and would be more than happy to share with second year team.


Now to answer your original question. Treads, while useful, can be pretty tough to get right, and are a lot harder than they seem. Use wheels.

[Roboticsismylif]

We have protyped the tank treads we would be using, and they're basically built. It's just we have been hearing some bad things about treads. We also do have a shooter on our robot to throw balls to the other side and make shoots if we have to.

[PayneTrain]

I also want to encourage you to know your limits, then push a little bit higher. I recommend that depending on your wheelbase, you go for a mix and match of AndyMark Plaction Roughtops and Omniwheels. If you only build a launcher and a hopper that allows for station loading and cross-court shots, you're doing a service to your alliance, especially if you can also align.

I'm not saying you shouldn't go for a strictly defensive robot even if that's your team's decision, but it's important to remember to not hold yourself back.

[pfreivald]

Originally Posted by Roboticsismylif

We have protyped the tank treads we would be using, and they're basically built. It's just we have been hearing some bad things about treads. We also do have a shooter on our robot to throw balls to the other side and make shoots if we have to.

Make sure you test them on carpet. Floor =/= carpet, and turning is often a problem in high-traction drive trains.

[Borobo]

I would really recommend against tank treads for 3 reasons:
1) They are more complicated than they appear. you will need the tread (fairly expensive) special wheels, and at least 8 wheels to do them well, and many teams use more
2)There is no large advantage over wheels. It is true that for a defensive robot they are good because they have immense traction, but they are also more inefficient that wheels. It is better overall to invest in a strong offensive robot, because the same charactersitics which make a robot good at offense also allow it to play at least some aspects of defense. For example, our robot this year has high traction mcmaster tread on 8 wheels and switching gearboxes, meaning that if for whatever reason we need to play defense, we will be able to simply push other robots around in low gear; maybe not as effectively as a tank treaded robot, but still enough to be useful in either role

[Kevin Ray]

We've done tank treads twice and they were awesome both times, but you have to have a lower pivot point in the middle of the track to aid in your turning. They also tend to be heavier overall than wheel systems--at least in our experience.
Finally, I agree...NEVER go into the build season planning to be a defensive bot. It's a recipe for failure. It's tantamount to playing basketball and only being defensive. Sooner or later the other guy(s) gonna score and you will be down points. You can almost always be a help by somehow getting balls to your allies so that they don't have to travel to the other side to get them. Be defensive only when everything else fails at the END of the build season--not in the beginning.

[Easton8]

From my experience treads are a real pain to use, and for a team that has never done them before I would definitely not recommend it. Building a basic 6WD from the kit of parts would be a lot simpler and save you a considerable amount of time and money. As for playing "defense" I would not recommend designing a robot for that, even if you think thats all you can build. To be an asset on the field you don't necessarily need an amazing robot that can score on the highest basket from half court every time. In fact a robot with just a decent shooter that stays on the other side of the field and feeds balls to other robots would almost as valuable an asset.

[Roboticsismylif]

We not actually using treads. We're planning on using a v-belt that has teeth on booth sides. We also are going with a rocker(spelling) bar like design. Kind of like a bulldozer. When we do turn the bot the treads turn in opposite directions. When this happens the rocker bar picks up the pulleys on the bottom ends leaving only the middle touching, reducing the amount of stress while turning.

[pfreivald]

Originally Posted by Roboticsismylif

We not actually using treads. We're planning on using a v-belt that has teeth on booth sides. We also are going with a rocker(spelling) bar like design. Kind of like a bulldozer. When we do turn the bot the treads turn in opposite directions. When this happens the rocker bar picks up the pulleys on the bottom ends leaving only the middle touching, reducing the amount of stress while turning.

That's pretty cool!