Not quite nanobots, since they're on the scale of 10 µm, rather than a nm, but still pretty tiny.
In a pair of small laboratories in Prague, a swarm of tens of millions of robots is being prepared, to be set loose en masse.
It is only fitting that here, in the town where the word robot was coined by author Karel Capek, the next generation of robotics should be envisioned.
But these won't be typical robots with gears and motors; they will instead be made of carefully designed chemical shells-within-shells, with receptors on their surface.
Instead of software and processors to guide them, their instructions will be written into the chemistry of their constituent parts. They are chemical robots, or as the 1.6m euro project's title has it, chobots.
In fact, notes Frantisek Stepanek of Prague's Institute of Chemical Technology, they are more like the robots described by Capek himself, formed of "...a blob of some kind of colloidal jelly that not even a dog would eat".
Dr Stepanek's robots will be small - tens of micrometres or less, a hair's breadth across - so that they can get into the tiniest places, or be dispersed in their millions for bigger tasks.
Those tasks will be to release a chemical payload, or mix two chemical reactants or "precursors" from different compartments within the chobots when they reach their goal.
Dr Stepanek sees them as one-time use chemical factories or repositories: cheap, expendable soldiers in the fight to, say, deliver medicines or seek out contaminants.
"They are synthetic single-celled organisms," he told visitors to his lab during the Research Connections 2009 conference.
Although they borrow ideas from life, he cautioned, they are not alive: they will neither evolve nor reproduce.
Smart move
They will be particularly well-suited for medicine, where they can be used for controlled delivery of a drug to the place where it is needed, rather than dosing the whole body.
"We would then be able use active molecules that are much more potent and therefore would have stronger side effects if they were applied in the standard way," he said.
Or they could be used in what he calls "intelligent cleaning", in which the robots could be released into a body of water, where they would seek out a source of contamination and upon finding one, chemically neutralise it.
For applications in medicine, allowing the robots to hitch a ride in the blood in order to reach their goal is one option.
But Dr Stepanek hopes to leverage recent advances in the mimicry of chemotaxis - the way that single-celled organisms move automatically toward a beneficial source like food or away from harmful ones.
What is more, besides a carefully chemically controlled locomotion, this kind of automatic response could form the basis of another idea borrowed from nature: swarming.
Swarming is a collective action of a number of agents, like ants or bees, though none of them has the master plan that describe what the swarm is doing.
It is a concept that has permeated the gears-and-motors robot community with increasing success in recent years.
"Each agent doesn't have the intelligence to know about the whole environment," Dr Stepanek says, "but by shared communication, there is a coherent, collective motion."
And how do robots made of chemicals communicate?
"A swarm will look for a microbe, say, and one locks onto it. It then releases a molecule that diffuses away," Dr Stepanek explained.
"It's not communication based on radio but on a molecule not normally present in the environment - an example of this in nature is pheromones."
The engineered chemotaxis takes care of the rest; the swarm marches inexorably toward the source of the signalling molecule, acting as a team.
Unique combination
The project will borrow heavily from other strands of research that are ongoing, but there are still significant challenges ahead.
Recent advances in materials science and polymer chemistry will help design the robots' membranes, semi-permeable barriers that let chemical payloads out or signalling molecules in.
They will be made either of tough silica - effectively sand shaped for the purpose - or hydrogels, squishy water-absorbing polymers that are not so unlike the membranes surrounding our own cells.
Biochemistry is still working out the details for the receptor molecules on the robots' surfaces so they can home in on their targets, and even more clever tricks will be needed to design the timely rupture of internal compartments that house the chemicals and precursors.
But Dr Stepanek has assembled a multidisciplinary team, saying that the combination of all these research strands in one place is unique in the world.
In truth, so far it's just a lot of chemistry in a dish and sketches on whiteboards, with some silica husks of the chemical robots nearly at the ready.
However, with the expertise in-house and 1.6m euros from the European Research Council, the team hopes to have its first functional prototypes swarming by the end of this year.