Flocking factors' assessment in case of destructive impact on swarm robotic systems

摘要

This paper dwells on exposures' analysis of open-environment operated swarm robotic systems. Flocking algorithms for swarm robotics, which use alignment and averaging of direction in local communication radius (after C. Reynolds' model) are considered. The problem where a small group of informed robots has to guide the swarm along a desired direction is also considered. Couzin I. D. et al. (2005) showed that even a minority of informed individuals is able to move the group toward the target Hereafter great number of swarm robotics' flocking algorithms were developed in order to increase swarm system's handiness. Several implicit leadership algorithms for distributed robotic systems were proposed. The idea of this paper is to study and simulate the operation of such kind of above mentioned algorithms in the situation when there are saboteurs which destructively impact swarm system. Specially developed by authors simulation software allows to vary the parameters of system operation, which can be rather the number of agents or their interaction features or etc. On the basis of the experimental data the authors produce the analysis of features in order to determine their relevance to the algorithm.