Whither microbots?

In the past few years there has been a great deal of discussion about the feasibility and desirability of developing microminiature robots which could be deployed en masse to do useful tasks. We attempt to examine some of the important issues involved in such a would-be endeavor, including matters of scale and controllability. Robots require considerable computational resources to be adaptive and flexible. Microscopic robots require no less, and this fact is highly problematic. In lieu of truly small robots, it would seem that the enlistment of naturally occurring or genetically modified organisms or biomolecular approaches would have distinct advantages. In the near term, and perhaps even very far term, it would seem that improvements in macroscopic-sized micro-motion robots and teleoperation systems which couple humans to such robots will be extremely important. In this regard, for example, scanned probe microscopy has made stunning progress in recent years in the imaging, analysis, and manipulation of mesoscale and atomic scale structures, yet the instruments themselves are very "macroscopic". There appears to be a promising role for MEMS as components for such systems. I present several examples from the research community that illustrate these themes, including work from our own laboratory on miniature modular robots for precision (micron level) assembly and high-fidelity haptic interface devices based on Lorentz magnetic levitation.