To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.build.mechaOpen lugnet.build.mecha in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Building / Mecha / 10475
10474  |  10476
Subject: 
Re: New Mecha - The Klymyth
Newsgroups: 
lugnet.build.mecha
Date: 
Thu, 25 Mar 2004 21:12:54 GMT
Viewed: 
1368 times
  
In lugnet.build.mecha, Kevin L. Clague wrote:
   Could you please educate me about “gait matrix”? I’ve not heard of this before. I’m really into making walkers and am always interested in seeing other’s works. Have you captured any of your experiments on a web page? I’d love to peek if you do.

Gait matrix is an old Robotics 101 term. It is representation of the temporal sequence of leg motions that a robot needs to walk without falling over. It gives the legs their coherent propulsive movement. Any autonomous walker needs to have this matrix encoded into it somehow (including living ones), otherwise it is just flailing legs around randomly, going nowhere.

My RCX robots have been wheeled because walkers tend to be limited in behavior to mostly walking, since that eats up the sensors and power outputs. Wheeled bots can have interesting behavior like my “Infradreadnaughts”(sic).

http://hometown.aol.com/Klaupacius/RCX/InfraDreadnaughts.html

My four legged walkers were pre-RCX and used an enhanced version of the old reciprocating leg technique, with the added features of moving knees and ankles driven by additional linkages. They had a very natural movement. I have no pictures, however. The gait matrix was encoded in the gearing linking all the legs together, driven by one motor (with a ton of gear reduction for torque).

  
I’ve made a lot of successful bipeds, and don’t consider ones that drag the feet on the ground “successful walkers”. Maybe my expectations are too high, or mapped inapropriately from biped to quad.

I use my creative juices to expand my abilities at articulated walkers (recently mostly using only pneumatics). I’m in the middle of a development cycle for a pneumatic hexapod. So far it can walk forward, just to prove to myself that a hexagonal body was viable for walking.

I agree though, I can hardly wait to see Eric’s creatures be more articulated.

I want to develop skills to bring out my abilities at putting stylistic skins over my creations. Eric and I are talking about combining efforts to merge his obviously highly developed artistic and mechanical skills with my skills at designing walkers (especially pneumatic ones), to see what the colaboration brings. I’m sure I’ll learn a ton of stuff from this effort.


Eric seems to stop building RCX control into his robots when he gets to the legs. Or he just stops building the robots at the torso! ;-) But he has the parts now for a walker. It’s just a question of the leg technology, and keeping the robot lightweight enough to move even with a stylish mecha skin.

K



Message is in Reply To:
  Re: New Mecha - The Klymyth
 
(...) I understand what you are saying. I think this depends on the vertical range of the gait. The higher rise of the feet, the more the quad tips when using three feet to balance. Could you please educate me about "gait matrix"? I've not heard of (...) (21 years ago, 25-Mar-04, to lugnet.build.mecha, FTX)

11 Messages in This Thread:




Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR