Deterministic and random partially self-avoiding walks in random media

Consider a set of N cities randomly distributed in the bulk of a hypercube with d dimensions. A walker, with memory μ, begins his route from a given city of this map and moves, at each discrete time step, to the nearest point, which has not been visited in the preceding μ steps. After reviewing the more interesting general results, we consider one-dimensional disordered media and show that the walker needs not to have full memory of its trajectory to explore the whole system, it suffices to have memory of order lnN/ln2.