Archive: 7 posts

---

First, i'll use the definition wolfdre made, to help people understanding the matter ! barycentre (plural barycentres) (mathematics, physics, astronomy) The point at the centre of a system; an average point, weighted according to mass or other attribute. The term is usually used in astronomy for the centre of mass about which a system rotates, for example, the moon and the earth rotate about a common point within the earth but not near the centre. Jupiter and the Sun rotate about a common point just outside the surface of the Sun. My question is : how to make a follower go to the barycentre of a 3+ magnetic tag system. In the experiments i've done, as long as there are more than two tags, it wont work like i would like it to

2013-01-22 09:55:00 Author: Unknown User

---

Hm... Do you mean the object should have equal distance to all tags? Or that "masses" thing like barycentric coordinates are usually meant? Untested idea: In the first case you can use (number of tags - 1) comparator circuits (direction combiner + positional sequencer essentially) to determine the closest tag (tag sensor measuring vicinity), and accordingly activate a 100% acceleration follower at 100 speed to flee from that tag; Just add a tolerance (maybe all tags +/- 2.5 units), or this is gonna break fast. Very fast. No warranty, but this idea might work. Also it may be a good-enough approximation for the second case most times. Hope that helps €dit: Follow the farthest tag instead; fleeing from the closest doesnt work too well. And at shapes w/ more than 3 corners this only partially works; also flat shapes caused problems in my test implementation.

2013-01-22 16:56:00 Author: Unknown User

---

Hey! Glad to see youve made it over! Ill look further into this at some point during the week. I have a suspicion that its going to take a bit of a work-around but then again im only speculating. Im totally swammped at this present moment but ill get back to you at some point during the week. Until then anyone out there is welcome to weigh in with their 2cents, (@Comphermc im lookin your way) this is a great question and sounds like fun to figure out. @Cendre: Is it possible you could post a mock-up of your logic for us? At least it gives us all an even starting point for debugging this particular scenario.

2013-01-22 19:31:00 Author: Wolfdre Posts: 218

---

This sounds like what Evret did for his 3D Rotation. Is it a case of revolving something around a changing point?

2013-01-22 21:42:00 Author: ballisticola Posts: 157

---

Following the farthest seem, and re-actualise seem to be the better solution, be still, it is an ugly solution ... Its not "clean" like we'd say in french. I Keep thinking about that, I'll post my solution in the end @wolfdre as i havent start yet the sackbot system, i haven't reallise anything yet. I ll make my limited checkpoints first Edit : 2 questions : 1) As there is a way to measure the distance between a magnetic tag and a tag sensor, is it possible, eventually, to activate a "mechanic pusher" (i dont know the english name) for 60% of his maximal range, with an input in it a 60% ? 2) is that possible to get negative numbers with soustraction ?

2013-01-22 22:08:00 Author: Unknown User

---

this may not be exactly what you are looking for, but it's a good place to start. the following setup will find the center of any number of tags based on the outer most tags, so it will not find center of tag mass (if you have 3 tags close together and one off to the right, it will only use the left most and right most tag to find center, and will ignore the fact the others are closer to the left etc) just set up a holo or sticker panel cross like in the pic below, have a tag sensor on each end with min range 0.1 and max set so it just reaches the sensor opposite. wire all four into an advanced mover set to 100% everything (see pic to see which is wired where) http://i1.lbp.me/img/ft/6313cdbc71b158609bd69130246b75713808a02e.jpg it detects the distance to the left and right most tags and the combiner forces the mover to move untill both distances match, same for up/down the main thing you need to remember while using this is that the effective area is roughly half the size of the cross, so you would need to enlarge the cross (sensors will scale up with it) so it is twice the size of your active play area. you can also increase the effective area by adding a few more sensors on each side and wire them into an OR b4 going to the combiner. another way that may improve it is to invert the sensors and the combiners, which would cause it to use the the outter part of the sensor range therefore the sensor arc wouldn't b as much of an issue. i came up with this a while back but as i havnt needed it myself I havnt really looked into the best way to implement it.

2013-01-23 07:45:00 Author: evret Posts: 612

---

Thank you very much Evret, if whatt i thought doesnt work, ill use your solution If its fonctionnal, it will work for every level, btw, so ill post about it. Just some math to deal with ! EDIT: Ok, thank you everyone, i did it, if someone needs it, tell me here so I make a tutorial !

2013-01-24 23:25:00 Author: Unknown User

Posts: 1077139    Threads: 69970    Members: 9661    Archive-Date: 2019-01-19

Aus dem Archiv wurden alle persönlichen Daten wie Name, Anschrift, Email etc. - aber auch sämtliche Inhalte wie z.B. persönliche Nachrichten - entfernt.
Die Nutzung dieser Webseite erfolgt ohne Speicherung personenbezogener Daten. Es werden keinerlei Cookies, Logs, 3rd-Party-Plugins etc. verwendet.