What type of drive train is the most maneuverable?

[PandaMan]

Although swerve drive can maneuver around objects, it is not the most physical of drive trains. From what I've seen, the most effective drive seems to be the 6WD.

[FourPenguins]

MORT's best drive train was a 4-wheel.
It used 5" (I think, I'm sure Brandon can correct me) Colson wheels with a very short, wide wheelbase (long robot, short wheelbase) to cut down on scrub. The tipping problems of a short wheelbase were solved with 4 ball casters, one on each corner. It handled beautifully, could get up some decent speed, and was a pretty vicious defender. (I'm pretty sure that robot provoked the no-wedge rule.) Keep in mind however, that this was a flatland drive train for triple-play.
General suggestions based on that bot:
*Colsons rock
*Find the happy balance in the ratio or use a 2speed tranny. (I don't remember what our ratio was.)
*Short, wide wheelbase reduces scrub.
*Ball casters are beautiful on flat land.
*Keep the game in mind. A flat game needs no clearance, a ramp does.

[MattLi]

Just an off topic post. If you want manuverable go with a wide robot. Doesn't matter how many wheels you have, it will still be better than a long robot.

[Madison]

Originally Posted by MattLi

Just an off topic post. If you want manuverable go with a wide robot. Doesn't matter how many wheels you have, it will still be better than a long robot.

All other things being equal, a typical six wheel drive along the long axis will have a shorter effective wheelbase than a four wheel drive arranged on the short axis.

[Joel J]

Is there a way to calculate the force (or torque, whatever) required to have a robot turn?

[Jack Jones]

Originally Posted by Joel J.

Is there a way to calculate the force (or torque, whatever) required to have a robot turn?

See this whitepaper:

[Joel J]

Originally Posted by Jack Jones

See this whitepaper:

I remember looking at that whitepaper. I recall that the only thing it said was that the length of the robot should be less that the wheelbase, for the robot to just begin to turn. I will look again to see if there is a method to determine how much force is required by a certain dimensioned robot, with a certain weight and a certain CoF with the carpet, to turn.

[edit: Ok, the equation is there. Great.]

[Richard Wallace]

Originally Posted by Joel J.

I remember looking at that whitepaper. I recall that the only thing it said was that the length of the robot should be less that the wheelbase, for the robot to just begin to turn. I will look again to see if there is a method to determine how much force is required by a certain dimensioned robot, with a certain weight and a certain CoF with the carpet, to turn.

[edit: Ok, the equation is there. Great.]

All you students out there, read the post above, at least twice. There is a great lesson in it.

Many of the references (books, articles, poems, CD threads, etc.) that you will encounter will have more to say to you when you re-read them a season or two later. The reference doesn't change from one reading to the next, but you do. And you get smarter, or more experienced, or both.

[Jay TenBrink]

I have read this white paper twice and I believe there is another factor to consider. The effective coefficient of friction in the lateral direction of a stationary tire is different (higher) than a tire that is slipping on the carpet.

We have had several successful skid steer robots with the wheelbase greater than the track width. These robots had large knobby pneumatic tires that were quite sticky. If the driver attempted to turn by energizing only the right side wheels and keep the left side stationary, the robot would not turn, as the white paper would suggest. If the operator pushed one stick forward and the other rearward with enough power to break the tires loose, the robot would turn. This was not a graceful turn, but worked well. This method requires the robot to have enough torque to break all wheels loose on carpet at the same time.

There is a good way to visualize this if you have ever driven on an icy road with a rear wheel drive vehicle. If you punch the accelerator and break the tires loose, the rear of the vehicle will move sideways as it slides down the crown of the road.

I cannot offer a formula to show this at this time, so I will have to leave the proof as an exercise for the reader.

Jay

[Jack Jones]

Originally Posted by Jay TenBrink

...I cannot offer a formula to show this at this time, so I will have to leave the proof as an exercise for the reader.

Jay

Dr. Joe's explained why here:

[Josh Murphy]

The drive train that the students came up with a couple of years ago is a pretty good system. We have used it in OCCRA for 2 years now and for the altering of our FIRST robots, which is four wheel drive in the back with casters or skyway wheels in the front. We also tried slides in the front instead of casters but I like the casters better. It is a simple machine for your budget and is very dependable.

[Dick Linn]

Ours was similar. The 4WD part had perhaps a 15 inch wheelbase and we used poly skids at the other end. Even so, the pneumatic tires were so grippy that steering was a bit herky jerky. Good system, though. Simple and reliable. he team won the Motorola Qualiity Award at the VCU regional for the overall 'bot.

[dlavery]

Originally Posted by M. Krass

All other things being equal, a typical six wheel drive along the long axis will have a shorter effective wheelbase than a four wheel drive arranged on the short axis.

Uhhhmmmmm, not so fast there. As a general statement, this is not true.

It all depends on how you are going to define a "typical six wheel drive" and what assumptions and caveats you are putting on that definition. I will maintain that in a "typical" six-wheel drive, all six wheels are co-planar. In that case, the wheelbase of the six-wheel drive is established by the Conservative Support Polygon, which is defined by the four outer wheels. As specified in the original statement, the wheels are arranged along the long axis of the robot. The four outer wheels determine the resulting wheelbase, which by definition is longer than a four-wheel drive arranged along the short axis of the robot.

This fundamental truism is only modified if the definition of a "typical" six-wheel drive is altered to promote non-standard configurations. For example, the common practice of moving the middle pair of wheels into a non-planar configuration. On a hard planar surface, this causes the robot to ride on the middle set of wheels and one of the "end" set of wheels. Only four wheels support the robot, not six. Thus, the wheelbase becomes approximately half of the long dimension of the robot. But this is actually no longer a true six-wheel mobility system. It is a set of two four-wheel systems that share a common pair of wheels (the center ones). In this case, you lose many of the advantages of a six-wheel drive (ie. distribution of weight across more points of contact with the floor, greater stability across the longer wheelbase, etc.).

-dave

[Madison]

Originally Posted by dlavery

Uhhhmmmmm, not so fast there. As a general statement, this is not true.

It all depends on how you are going to define a "typical six wheel drive" and what assumptions and caveats you are putting on that definition. I will maintain that in a "typical" six-wheel drive, all six wheels are co-planar. In that case, the wheelbase of the six-wheel drive is established by the Conservative Support Polygon, which is defined by the four outer wheels. As specified in the original statement, the wheels are arranged along the long axis of the robot. The four outer wheels determine the resulting wheelbase, which by definition is longer than a four-wheel drive arranged along the short axis of the robot.

This fundamental truism is only modified if the definition of a "typical" six-wheel drive is altered to promote non-standard configurations. For example, the common practice of moving the middle pair of wheels into a non-planar configuration. On a hard planar surface, this causes the robot to ride on the middle set of wheels and one of the "end" set of wheels. Only four wheels support the robot, not six. Thus, the wheelbase becomes approximately half of the long dimension of the robot. But this is actually no longer a true six-wheel mobility system. It is a set of two four-wheel systems that share a common pair of wheels (the center ones). In this case, you lose many of the advantages of a six-wheel drive (ie. distribution of weight across more points of contact with the floor, greater stability across the longer wheelbase, etc.).

-dave

Yes, this is entirely true -- and what I get for posting at 3 AM or some ungodly hour like that. In FIRST, I consider a 'typical' 6WD arrangement to include a lowered center wheel, but I can see that in another application, this definition would be incorrect. Trucks or airplanes, for example, may ride on six wheels rather four to, as you mention, better distribute their weight across a surface. I meant to speak only with the confines of FIRST use -- and even then, perhaps it's erroneous to assume that most consider the 'typical' 6WD arrangement to have a lowered center wheel.