Evolution of Direct Communication for a Swarm-bot Performing Hole Avoidance
Evolution of Direct Communication for a Swarm-bot Performing Hole Avoidance
复制标题
群机器人执行漏洞规避的直接通信的演变
DOI:
--
复制
发表时间:
2004
期刊:
影响因子:
--
通讯作者:
M. Dorigo
中科院分区:
文献类型:
--
作者:
V. Trianni;T. H. Labella;M. Dorigo
Communication is often required for coordination of collective behaviours. Social insects like ants, termites or bees make use of different forms of communication, which can be roughly classified in three classes: indirect (stigmergic) communication, direct interaction and direct communication. The use of stigmergic communication is predominant in social insects (e.g., the pheromone trails in ants), but also direct interactions (e.g., antennation in ants) and direct communication can be observed (e.g., the waggle dance of honey bee workers). Direct communication may be beneficial when a fast reaction is expected, as for instance, when a danger is detected and countermeasures must be taken. This is the case of hole avoidance, the task studied in this paper: a group of self-assembled robots – called swarm-bot – coordinately explores an arena containing holes, avoiding to fall into them. In particular, we study the use of direct communication in order to achieve a reaction to the detection of a hole faster than with the sole use of direct interactions through physical links. We rely on artificial evolution for the synthesis of neural network controllers, showing that evolving behaviours that make use of direct communication is more effective than exploiting direct interactions only.