(Sorry: too late to be read, noprocrast and comment posting timeout conspired against me.)
I think the interesting part of this is the gait mechanism which is
unaddressed in the article.
A little experimentation with my fingers on the desktop confirms a
single actuator to move the legs between two geometries, and the
ability to turn suction on and off to the pads is sufficient to
climb walls. (Turning probably requires that you can individually
activate the pad suctions and have some flex available.)
time ====> repeat
- O O
\ \ \
x x x
|\ |\ |\
- O O | O | O | -
/ / / | | |
x x x | - | O | O
/| /| /| |/ |/ |/
O O | O | - | x x x
\ | | | / / /
x - | O | O | O O -
|\ \| \| \|
| - x x x
| \ \ \
| O O O -
|/
x
/
-
odd toggle all even toggle all odd
only position suck only position suck only
suck actuator suck actuator suck
if battery weight is a problem, can't they attach a external power cord to it ? I mean it's not like the water pressure is coming from inside the robot.
I think the interesting part of this is the gait mechanism which is unaddressed in the article.
A little experimentation with my fingers on the desktop confirms a single actuator to move the legs between two geometries, and the ability to turn suction on and off to the pads is sufficient to climb walls. (Turning probably requires that you can individually activate the pad suctions and have some flex available.)