Thursday, May 29, 2008

Building a Better Wall Climber

Continued from page 1

By Duncan Graham-Rowe

These charges are similar to the electrostatic charges generated when a person walks across a carpet, says Prahlad. The difference is that with the robot, the charges are generated by a power supply, and the electrodes are well insulated so that the field strength between the electrodes is great. What's more, there is no neutralization, or grounding, of this charge when contact with the wall is made (and there is therefore no spark).

Individually, the attractive forces between opposite charges within the robot and the wall are tiny, says Prahlad. But cumulatively, they are sufficient to hold the robot to the wall.

In its current form, the robot, featuring off-the-shelf components, moves pairs of tracks, like a tank, in which each tread is made up of strips of flexible sheets containing the pairs of electrodes. These sheets look a bit like Post-it notes and make contact with the wall as the tread moves around, peeling off again as the tread leaves the wall. However, this design is merely for convenience, says Prahlad. The beauty of electroadhesion is that you don't have to peel the pad off, which can't be said for the geckolike adhesion approach.

The flip side to this, says Pugno, is that you need a power supply to maintain adhesion. If your power supply fails, then so does your stickiness. Even so, what is most remarkable about this approach is just how little power is needed to maintain adhesion. For although 5,000 volts are applied to the electrodes, a current of just 50 nano amps drives this, meaning that only microwatts of power are used. The vast majority of the power is spent driving the motors that turn the robot's sticky, tank-like tracks. In theory, if the same technology were used to stick a picture on a wall, a small battery could keep it hanging for decades before running out.

Prahlad says that some of the robotic applications are for military purposes, such as to allow robots to set up ad hoc networks by scaling structures in urban or rural battlegrounds, or for surveillance applications.

Electroadhesion is not a new concept and has even been proposed as a means of lifting silicon wafers during chip manufacturing, says Sitti. "However, its application to climbing robots is new." What would be interesting now is to see how a hybrid approach works using setae-like fibers that also have electroadhesive properties, he says. This could potentially provide the robot with a stronger grip.