Jansen Mechanism

After seeing the recent revival of the popular Theo Jansen Mechanism video.

http://www.grasshopper3d.com/video/rhino-grasshopper-kinematic-system

I thought I would finally get around to having a go myself.

The First file uploaded below was a crude attempt to blindly reconstruct the animated gif shown below from http://www.tedngai.net/experiments/jansen-mechanism.html

The second File is a more refined attempt after finding the "magic" Values illustrated below:

These were obtained from:

http://www.instructables.com/community/Theo-jansen-mechanism/

If anyone finds this useful please enjoy!

Attachments:

JansenMech.gh, 38 KB
TheoJensenMechanism.gh, 68 KB

Replies to This Discussion

Permalink Reply by Danny Boyes on May 12, 2014 at 12:23pm

*** U P D A T E D ***

Permalink Reply by Nick Tyrer on May 12, 2014 at 3:00pm

haha how random Danny. I just built one of those the other day from a kit. And as soon as i had finished i built that same mechanism in gh from the same instructables page..

Permalink Reply by Danny Boyes on May 13, 2014 at 3:39am

Very cool! (or was the hair drier set to warm :))

Permalink Reply by Nick Tyrer on May 13, 2014 at 4:05am

ooo danny, that is terrible! made me cringe. haha, if you are interested in a kit. I bought that one on Ebay, from China for £10 (including delivery). I wanted to get some of the other kits but the price jumps considerably for the other types of walker

Permalink Reply by Danny Boyes on May 13, 2014 at 4:11am

I believe my taxi for one is here...

Permalink Reply by peter fotiadis on July 3, 2014 at 2:51am

The thing anemonized.

PS: Anemone could change the way that people do stuff in GH ... but a "little" thing is omitted (see my last reply to find why : http://www.grasshopper3d.com/group/anemone/forum/topics/help-with-l...)

Attachments:

JansenMech-anemone.gh, 66 KB

Permalink Reply by Mateusz Zwierzycki on July 3, 2014 at 4:07am

Nice def here, I wrote you a response in the anemone group discussion... I think I didn't get the info about this discussion in the gh forum newsletter... ;/

Permalink Reply by Marios Tsiliakos on July 3, 2014 at 4:27am

Nice,I just saw this thread.

I wrote a paper with some colleagues on this back in 2009, and re-created the evolutionary solver that Theo Jansen must have used, in Processing. I got slightly different magic numbers, maybe local optimas, maybe I did something wrong.

I 'll try to dig up my files in case I find the report, videos (one of them actually displays the phenotype of the computer program Theo Jansen run) and a thesis that recreated the whole process by an engD candidate in S. Africa.

To find the "magic" numbers is for me the most exciting part of this process, rather thatn the kinetic structure its self.

Permalink Reply by peter fotiadis on July 5, 2014 at 8:43am

Well...I've spend some time more on that thing:

1. We need 3 combos of "legs" in (at least) 2 sets - task for the kinematic freaks. Of course there's always the "other way" : what about this? (it could fly with a Porsche flat six).

2. Added a double (nested) Anemone thing (and the Mateusz version) and some comments.

3. Added a stupid "arm maker" cluster ... primitive/ugly/pointless - see one prototype attached about how to do it (you'll need a top feature driven CAD app for this - notice the Teflon low friction ring).

4. In order to "adapt" the cluster arm you need some "stretch" capability (orient, scale et all are the 1st step). Of course putting the cluster into the 2*loop is the art of pointless (Mateusz misinterpreted my bitter comments as regards the "slow" thing, I had absolutely no intension to recreate "live" the arm).

be the Force (the dark option) with you all.

Attachments:

JansenMech-anemone-nested2.gh, 132 KB
Arm-ASSE-V4.zip, 3.8 MB

Permalink Reply by peter fotiadis on July 6, 2014 at 1:00am

Another hour spend on that thing:

1. Arm maker cluster replaced with a ... worst arm maker (that's progress I guess). The odd thing is that GH still can't bake the arm out of the cluster (meaning the obvious) ... whilst he can bake the "very" same thing after the orient and/or the scale component. That's the mysteries of CAD friends, he he.

2. As regards putting the "solid" Arms around (some z offset is required, obviously) ... well... the "transformation" logic (after the orient thing) is wrong since all the knobs have the same size (and same bolts: i.e. same diameter holes) ... only a stretch could do it (but I confess that I can't get the gist of that Rhino command - even Microstation beats Rhino on that matter hands down).

3. NEVER attempt to put the cluster (that stupid Arm Maker) inside the Anemone loop - can you guess why?

Attachments: