Exhaustive Geographic Search with Mobile Robots Along Space-Filling Curves

Swarms of mobile robots can be tasked with searching a geographic region for targets of interest, such as buried land mines. We assume that the individual robots are equipped with sensors tuned to the targets of interest, that these sensors have limited range, and that the robots can communicate with one another to enable cooperation. How can a swarm of cooperating sensate robots efficiently search a given geographic region for targets in the absence of a priori information about the targets' locations? Many of the "obvious" approaches are inefficient or lack robustness. One efficient approach is to have the robots traverse a spacefilling curve. For many geographic search applications, this method is energy-frugal, highly robust, and provides guaranteed coverage in a finite time that decreases as the reciprocal of the number of robots sharing the search task. Furthermore, it minimizes the amount of robot-torobot communication needed for the robots to organize their movements. This report presents some preliminary results from applying the Hilbert space-filling curve to geographic search by mobile robots.