6

enter image description hereenter image description here

I am trying to create a tricopter which will be used for SLAM, but my main problem is that before starting to construct it I am not sure whether or not it will work at all. The problem is that I need a sensor to be rotating on an axis vertical to those of the rotors. Moreover that part will not be on top or below the drone but rotating around its "chassis axis"(blue line on second ppicture), because the sensor must not have anything blocking its view. Is it possible to achieve it? If yes what is the optimum placement of the motors and the moving sensor? Since English is not my native language I understand that some things may not make sense please feel free to ask me anything you don't understand.

  • can't you do it like that : https://goo.gl/4tB7vO ? But that's not the problem I am trying to solve. My problem is the roll i think –  Dec 09 '16 at 15:39
  • @PanteleimonStamatakis this may help, tell me if it does and I'll put it as an answer with the key bits in it. https://books.google.co.uk/books?id=-OjZBQAAQBAJ&pg=PA20&lpg=PA20&dq=Is+it+possible+to+achieve+torque+equilibrium+in+order+to+fly?&source=bl&ots=IqH9P7GSKh&sig=Hg9JzBCy00oDuUTUhDvgwokWVOk&hl=en&sa=X&ved=0ahUKEwj4p_-OuufQAhUmAcAKHXSQD3QQ6AEINDAE#v=onepage&q=Is%20it%20possible%20to%20achieve%20torque%20equilibrium%20in%20order%20to%20fly%3F&f=false – anonymous Dec 09 '16 at 15:46
  • You have your answer in the comments as far as I am concerned (don't know why they didn't 'answer'. Anyway, I would change the graphic as I don't understand it. – user6035379 Dec 09 '16 at 16:03
  • @anonymous i know what that link was about, the thing is, how do i achieve it? –  Dec 09 '16 at 16:48
  • Is there a reason this couldn't be built based on a more conventional quadcopter? – Eugene Styer Dec 09 '16 at 18:34
  • @mins Google search is at least a little better than you think; it counts matches with words like "is" or "to" very little, precisely because they are so common. The bigger problem with this search is that the book in question often mentions "equilibrium" in ways that have little or nothing to do with a tricopter. – David K Dec 10 '16 at 18:19
  • @mins My reference is that my classmates in grad school were working on this problem over 20 years ago, before Google existed, and even then we knew that matches on some words are stronger indicators of relevance than matches on other words are. Some common words are not counted at all by Google. – David K Dec 10 '16 at 19:23
  • @DavidK: The search in question is within Google Books, rules are a bit different here. For example "the", "for", "to" are perfectly valid and taken into account. I can search the expression "one of the contributors to this volume" using only "one, "the, "to" and the wildcard. But as you can see other pages containing only one of the keywords, including "the" will also be found and ranked below the page that contains all words. – mins Dec 10 '16 at 19:50
  • @mins The page I pointed to is a simplified explanation; the precise algorithms are proprietary. There is some natural language understanding involved, and words that are not in the query (but are considered similar) may be counted. As you note, the algorithms are also modified for different kinds of "pages" that are searched. Your example shows that your search string is somehow interpreted to yield some "hits" and that any words in the search are highlighted in each block of text returned. It doesn't tell us why those particular blocks of text were chosen. – David K Dec 10 '16 at 20:14
  • 1
    @mins In any case, I think we agree the original query was not very successful; I just think the problem is the inability of the search algorithm to clearly identify the concepts meant by the "big" words in the query, not the presence of a few "little" words. For example we get a hit at the end that has only two highlighted words, far fewer than many pages that were skipped, but it has many words such as "angle", "rotational", "force", and "inertia" that are associated with "torque". – David K Dec 10 '16 at 20:18
  • Is there a specific reason to make a tricopter, not a quadcopter? An even number of rotors allows to cancel out torques, a greater number of rotors than 3 allows for more degrees of freedom in trimming the craft for CoG shifts. – Koyovis Nov 17 '17 at 15:40

1 Answers1

2

If I understand the problem correctly, there will be a camera attached around the perimeter of the copter, which must have an uninterrupted view.

If the camera just swings around a perimeter, there will be two problems:

  • Trim of the device. There are only three rotors. If the device is in a trimmed hover and the camera swings around, the device needs to be re-trimmed rapidly to prevent roll/pitch.
  • Reaction yaw. The device starts spinning the other way that the camera is swinging.

Best way is to give the camera a fixed place and make the whole device spin, via gimballing the main rotors or using two vertical propellers. A counter-weight to the camera will put the CoG in the centre.

Koyovis
  • 61,680
  • 11
  • 169
  • 289