I am simulating a wheeled robot of six-wheels and can be independently steered, like MER-Opportunity. The wheeled robot can perform throttling forward,

||---|| <--wheel orientation
||   ||


//---// <--wheel orientation when heading is 45
//   //

and turning on the spot.

//---\\ <--wheel orientation
||   ||

My question is: Is it correct to say that I have 2 motion primitives? Throttling forward is basically crab-motion with heading zero.

  • $\begingroup$ hmm, I suppose it depends if orientation is important... $\endgroup$ Oct 15 '14 at 9:00
  • $\begingroup$ Well, orientation is important. $\endgroup$
    – ikel
    Oct 16 '14 at 9:50

While I am not really sure what you exactly mean by "motion primitive", having the ability to do the "crab-motion" enables you to translate regardless of your orientation and without changing it (assuming 180 degrees of steering) which would not be the case if you had only one pair of steerable wheels (like in cars, remember the parallel parking problem).

Note however that this is still non-holomonic motion system unless the wheels are omniwheels (you cannot rotate and translate independently).

  • $\begingroup$ Motion primitive (link) in the sense of a small action that a platform is able to perform. Normally, a complex motion is not suitable to solve all problems, rigid. Motion primitives can be used in series of execution to produce a complex motion to solve a problem. Hmm, I'm not sure if I explained it correctly. $\endgroup$
    – ikel
    Oct 16 '14 at 9:54
  • 1
    $\begingroup$ Then I guess it depends on the question, is your platform oriented (in sense that it has front and back, regardless of current wheels orientation) or not. If it's not then I'd say you have 2 motion primitives. If it is oriented (because for example you have some camera at the front or something) then I'd say you have 3 motion primitives (rotate wheels, rotate robot, translate in the direction of wheels). $\endgroup$ Oct 16 '14 at 15:05
  • $\begingroup$ Voted as best answer as you made mention about non-holomonic moation system and platform oriented, which helped a lot. $\endgroup$
    – ikel
    Oct 17 '14 at 19:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.