A new strategy is proposed for shape assembly of robot swarms based on the idea of mean-shift exploration: when a robot is surrounded by neighbouring robots and unoccupied locations, it would actively give up its current location by exploring the highest density of nearby unoccupied locations in the desired shape. This idea is realised by adapting the mean-shift algorithm, which is an optimisation technique widely used in machine learning for locating the maxima of a density function.
When surrounded by other robots and unoccupied locations, a robot explores the highest density of nearby unoccupied locations in the desired shape, as identified by the mean-shift optimisation.
The proposed strategy was verified by experiments with swarms of 50 ground robots, which demonstrates its potential to be adapted to generate interesting behaviours including shape regeneration, cooperative cargo transportation, and complex environment exploration.
The study titled, "Mean-shift exploration in shape assembly of robot swarms", has been published in Nature Communications. Dr Roderich Gross at the University of ºù«Ӱҵ has collaborated on this work with researchers from Westlake University, Beihang University and Tsinghua University. The paper can be accessed via the following URL: