Probabilistic behaviour in ants: A strategy of errors?

Abstract

Aiiiinnl hchaviour is probahilistic. This is exemplified by the comnninicalioii bchîU'iour of nuls diiriiig fooil-searcfiing. Expérimental évidence deinonsiratcs lliat speeies difïer in the accuracy of tlicir rccruitmenl. W e show hcrc. with the hclp of a vcry simple mathematical model, that the randomncss of behaviour can have an adaptative advantage for ants. The model demonstraies that the degree of randomness eould be optimally "luned" to pnrticulnr ecological conditions, siich as food quantity and distribution. Society is to try to eliminate external fluctuations. One way is to damp the external fluctuations with hoineostatic mechanisms: the colony tries to

Random behaviour, amplification processes and number of participants: How they contribute to the foraging properties of ants

“…With more than one source, however, the lost ants enable the discovery of other sources and there is an optimal error level which minimises the global exploitation time. There is also a sub-optimal solution, and an error level above which collective exploitation is not possible [19,20]. Some inacurracy during mass recruitment could therefore be functional when food sources are aggregated, and at least in theory the level of accuracy could be tuned to the natural distribution of the food sources.…”

Bio-Inspired Approach for Autonomous Routing in FMS

“…The first experiments related to the industrial use of stigmergy were conducted in the early 1980s by Deneubourg et al (1983), who simulated "ant-like robots". Since then, many researchers (e.g., Ferber, 1995;Arkin, 1998;Dorigo & Colombetti, 1998) have applied this concept when studying robot collectives and working to solve optimization problems (e.g., Travelling Salesman Problems, Network Routing for telecommunications and the Internet).…”

“…Our approach to FMS is based on the behaviour of biological systems, such as ant colonies (Deneubourg et al, 1983;). An FMS can be seen as a network of nodes that are interconnected by uni/bi-directional paths on which mobile product entities navigate.…”

Error, communication and learning in ant societies

“…With more than one source however the lost ants enable the discovery of other sources and there is thus an optimal error level which minimises the exploitation time. There is also a sub-optimal solution, and an error level above which collective exploitation is not possible (Pasteels et al, 1982;Deneubourg et al, 1983).…”