[archiver]

Posted by Dodd Stacy.

Engineer on team #95, Lebanon Robotics Team, from Lebanon High School and CRREL/CREARE.

Posted on 9/7/2000 12:42 PM MST


In Reply to: Holonomic warnings posted by Joe Johnson on 9/7/2000 10:44 AM MST:




: My one issue with holonomic drive wheels (also known as omni-rollers or omnidirectional wheels) is that in the direction transverse to the 'major' direction of rolling, the effect wheel diameter gets pretty small.

: I know that it can be made to work, but I still get a bit nerveous with wheels smaller than 3 inches or so. The holonomic wheels I have seen are much smaller than that when rolled against the grain.

: Just my two cents worth.

: Again, the idea is very clever and has some very cool advantages. I am only pointing out that there are some issues to be aware of when you are making your design decisions.

: Joe J.

Your point is apt, at least when considering the type of wheels most folks have seen, the so-called Mecanum or roller-rim type. We built this type last year, with 8 staggered rim rollers of 1 1/2' diameter on a 5' diameter wheel. We thought about the limited transverse obstacle climbing ability you note, decided the floor was pretty flat (and the rules precluded laying down tethered obstacles, etc - remember Torroid Terror?), and tried it. It worked pretty well, and was awfully handy in maneuvering the puck around the field after clamping onto the pole. Also for positioning TO clamp on the pole.

Anyway, I'm responding to your message not to be argumentative, but to point out that there are other classes of holonomic/omni-directional 'wheels' out there with different attributes, and a group of minds like the FIRST community can come up with more yet. In developing an omni-directional material handling platform at work for use on aircraft carrier decks, we came across (and developed further) a family of designs that uses a large diameter sphere as the rolling element in contact with the deck/floor. This 'wheel' confronts any obstacle with its major diameter, regardless of direction of motion. This is handy for rolling over 2' diameter arresting gear cable, tie-down chains, etc.

All of the rollers, bearings, and other complex bits for loading the weight of a bot onto the sphere, keeping the sphere corraled in the bot, enabling it to rotate in all directions, and driving the sphere in 1 or 2 directions - they all bear from above upon the nice predictable surface of the sphere (as long as you keep it swept clean). These 'wheel' systems have more parts and are more mechanically involved than the Mecanum type, but they finesse the concern you raise.

Design decisions always involve trade off judgements, and there are always more ways to do things than we already know about. I'm trying to embolden Josh and other teams like his to tackle the manageable challenges of omni-directional drive now and field creative wonders to thrill and delight us all in 2001. Actually, I'd like to see some team go Dean's IBot one better and make a bot that balanced on ONE spherical omni-directional rolling element. Kind of a unipsycle.

Dodd