Computing with the Collective Intelligence of Honey Bees – a Survey
Total Page:16
File Type:pdf, Size:1020Kb
This document is downloaded from DR‑NTU (https://dr.ntu.edu.sg) Nanyang Technological University, Singapore. Computing with the collective intelligence of honey bees – A survey Rajasekhar, Anguluri; Lynn, Nandar; Das, Swagatam; Suganthan, P. N. 2016 Rajasekhar, A., Lynn, N., Das, S., & Suganthan, P. N. (2017). Computing with the collective intelligence of honey bees – A survey. Swarm and Evolutionary Computation, 32, 25‑48. https://hdl.handle.net/10356/83447 https://doi.org/10.1016/j.swevo.2016.06.001 © 2016 Elsevier. This is the author created version of a work that has been peer reviewed and accepted for publication by Swarm and Evolutionary Computation, Elsevier. It incorporates referee’s comments but changes resulting from the publishing process, such as copyediting, structural formatting, may not be reflected in this document. The published version is available at: [http://dx.doi.org/10.1016/j.swevo.2016.06.001]. Downloaded on 28 Sep 2021 17:07:28 SGT 1 Computing with the Collective Intelligence of Honey Bees – A Survey Anguluri Rajasekhar, Nandar Lynn, Swagatam Das, and P. N. Suganthan Abstract –Over past few decades, families of algorithms based The term swarm is used for an aggregation of natural creatures on the intelligent group behaviors of social creatures like ants, like fishes, birds and insects such as ants, termites, and bees. birds, fishes, and bacteria have been extensively studied and used The members in the swarm interact with each other without for computer-aided optimization. Recently there has been a surge any central supervision and they move based on the of interest in developing algorithms for search, optimization, and information obtained from their neighbourhood. In this paper, communication by simulating different aspects of the social life of a very well-known creature: the honey bee. Several articles a comprehensive survey of the computing algorithms based on reporting the success of the heuristics based on swarming, the social behaviours of honey bees is presented including mating, and foraging behaviors of the honey bees are being different variants of the bee-inspired algorithms and published on a regular basis. In this paper we provide a brief but applications of those variants to real-world optimization comprehensive survey of the entire horizon of research so far problems. undertaken on the algorithms inspired by the honey bees. The paper is organized in the following way. Section II Starting with the biological perspectives and motivations, we discusses the biological perspectives of honey bees and the outline the major bees-inspired algorithms, their prospects in the motivations behind the algorithms inspired by different respective problem domains and their similarities and activities of honey bee swarm. Section III outlines various dissimilarities with the other swarm intelligence algorithms. We also provide an account of the engineering applications of these algorithms based on the mating and marriage of queen bees. In algorithms. Finally we identify some open research issues and Section IV, studies based on developments of methods promising application areas for the bees-inspired computing involving foraging and communications between bees are techniques. reviewed. Section V provides an overview of the methods Keywords-Nature inspired computing, swarm intelligence, honey based on honey bee swarming. In Section VI algorithms based bees foraging, particle swarm optimization, bee colony optimisation; on spatial memory and navigation methods are discussed. Section VII presents a comprehensive outline of the I. INTRODUCTION algorithms based on the division of labor in honeybees. EES are four-winged flower-feeding flying insects having Finally, conclusions are drawn in Section VIII with a Bsimilarities with ants and wasps and they belong to the discussion of the potential future research issues. A summary family of Apoidea. They are well known for their pollination of the key concepts of some most representative bee-inspired and production of honey and beeswax. There are around algorithms along with the meaning of various metaphors 20,000 known species of bees organized in 7 to 9 recognized commonly found in the related literature is provided in Table families. Among all the categories of bees, bumblebees and 2. honey bees have a very peculiar life cycle which has attracted several researchers to study about the facts associated with II. HONEY BEES – BIOLOGICAL PERSPECTIVES AND them. These studies led to different theories which are highly MOTIVATIONS suitable to meet the needs of engineering applications. In a In nature honey bees create nests called ―hives‖ (consisting honey bee swarm, the tasks are dynamically assigned and the up to 20,000 individuals) and they work together following a bees adapts to the new environment in a collaboratively highly structured social order. The brief taxonomy involved in intelligent manner [S1, S2, S3]. the life cycle of bees is shown in Fig.1. The following section In the recent past, with the computational cost having been provides a brief description of the evolution of bees and their almost dramatically reduced, the researchers have drawn the role played in collection of nectar. ideas from nature to solve difficult computational problems. Among the population-based models for computer aided 1. Honey Bee Hive problem solving, the insect colonies exhibit a significant level of interaction through mutual co-operation and organizational This is the place where new bees take birth. The hive behavior which are necessary for survival in the life process. consists of a number of vertical combs. There are two kinds of Computer scientists observed that efficient metaheuristic comb cells; the worker cells and the drone cells. In due methods can be developed, based on the honey bees‘ course of time occasionally the bees build a third type of cells, intelligent behaviors of co-operation, allocating the task force the queen cells, in which the queen bees are reared. and finding the food source using the mutual interaction as guidance. Note: Due to space limitation the supplementary reference list cited as [Sxxx] will not be published in print, but as an on-line document only. The list is appended with this version. 2 i)Queen Bee (source) she should visit. At the end foragers carry the In general ―Queen Bee‖ refers to an adult mated female. collected nectar to hive. Her sole purpose in life is to lay eggs and to produce more ii) Communication: bees. She is the only female in the hive. A bee returned to the hive from a good food source ii) Drone Bees performs a kind of ‗dance‘ on the vertical comb. In Drones are the male honey bees developed from eggs that general these are given particular name based on the have not been fertilized. Drones mate with a queen in front profitability and distance of the flower (source). For of the hive through the phase of mating flight. instance, if the distance is beyond 150 meters the bee performs waggle dance, a particular figure eight dance to iii) Worker Bees share the information regarding the direction and distance Workers are the most populated non-mating female bees, of the food sources. Other workers follow the dancing bee constituting over 98% of the colony‘s population. The and then learn the direction and distance of the food worker bees play a major role in collection of nectar and in source based upon the dancing behavior. A waggle dance safe-guarding the hive from potential intruders through with a very short waggle run used to be characterized as a social interactions. Several bee colony based algorithms round recruitment dance. Forager honey bees of the imitate the collective intelligence shown by the worker species Apismellifera can perform tremble dance to bees. recruit more receiver honey bees to collect nectar from the workers. Tremble dance resembles the waggle dance, however, a forager uses it when the foraging bee senses a long delay in dumping its nectar or a dearth of the receiver bees. 3. Honey Bee Swarming and Nest-site-Selection Swarming is often considered to be the process in which bees abandon their hives and collectively move to perform various life-supporting activities. One situation includes rapid reproduction of honey bees in the colony resulting in the increase of population. The reasons may include abundant availability of nectar and favorable environmental conditions. On other time it may happen that the honey flow has been restricted or ailments which forces the worker bees to leave the hive. As soon as they emerge as a swarm they aim to construct a new site after deciding the most suitable site. The most experienced scout bees search a large area for a suitable new site. Bees that have found good sites return to the cluster and the strength of their waggle dance becomes proportional Fig.1.The taxonomy involved in the life cycle of honey bees (in some sense) to the quality of the nest-site [S10, S11, S12]. 2. Honey Bee Foraging and Communication 4. Spatial Memory and Navigation Learning plays a vital role in foraging. Von Frisch and Often foragers (worker bees) come across a situation where Seeley are considered to be the pioneers in studying the finding a way between food source and hive becomes learning capabilities of honey bees. From different extremely difficult due to fractal oriented landscapes and experiments and many studies conducted by Von Frisch [S4, longer distances. Honey bees usually overcome this problem S5], Von Frisch and Lindauer [S6], Seeley [S7], Chalifman with the help of landmarks and celestial cues. According to [S8] and Lindauer [S9], it is observed that a single forager will Menzel et al. [S13], foragers use a map oriented organization visit the different flowers (food sources) in the morning. If regarding the spatial memory, which is based on the there is a particular kind of flower which has sufficient computations of land marks and celestial cues [S14, S15].