Subject:
|
Designing a bipedal robot leg
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Mon, 20 May 2002 21:27:27 GMT
|
Original-From:
|
PeterBalch <peterbalch@STOPSPAMcompuserve.com>
|
Viewed:
|
617 times
|
| |
|
|
Andy
> If you were to design a leg for a bipedal robot, how would you do it? • Would
> it have a knee, or a moveable ankle?
Both.
Many years ago, in the days before RCXs, I constructed a few walking bipeds
with knees.
I found the problem with swinging a large weight around to maintain balance
was that the momentum of the weight often toppled the robot.
My final, very successful, design tilted the ankles from side to side so
the centre of gravity stayed over the foot on the ground. The body was 12cm
x 5cm x 5cm and was carried 10cm above the ground. Each foot was a 5x10
stud flat plate. Three 40T gears on each side provided (a) front-back drive
for the leg (b) flex-extend of the knee (c) tilt of the ankle. Getting the
phases just right took a while.
The problem with all such knee/ankle designs is steering. I never made a
successful steering biped. The modern designs which twist the legs solve
the problems very neatly indeed.
> Would you use pneumatics, motors and
> cables, or the various muscle wires or pneumatic muscles available?
Motors. It's too hard to keep everything in the right phase using Lego
pneumatics.
> Would shortening the distance between the legs
> help in this?
Do you mean the side-to-side distance between the legs? Yes. The closer
they are together the easier it is to transfer the weight from one to the
other. All bipedal walking animals keep their feet as close to the centre
line as they reasonably can. Only bipedal hopping animals keep their feet
apart.
As an aside: I also attempted 6-legged walkers. I initially assumed that
they should be steered by taking longer or shorter strides on the left or
right sides. It doesn't work at all well. The design assumes that the front
and rear feet will slip sideways. However, the forces generated by the
longer strides are insufficient to overcome the sideways friction. The feet
slip but forwards and backwards - not side to side.
My eventual successful design used Ackerman sterring - like a car's front
wheels. The front legs of the walker could turn to one side of the other.
It's tricky to get the drive to then while they are turning; the drive
shaft must be the same "kingpin" that each leg steers on. Ackerman sterring
at both the front and rear would be even better.
Peter
|
|
Message has 1 Reply:
 | | Re: Designing a bipedal robot leg
|
| My comments are interspersed, with no delimeters. I hate OE. :P ----- Original Message ----- From: "PeterBalch" <PeterBalch@compuserve.com> Cc: "[unknown]" <lego-robotics@crynwr.com> Sent: Monday, May 20, 2002 5:27 PM Subject: Designing a bipedal (...) (22 years ago, 21-May-02, to lugnet.robotics)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
