Swarm Intelligence: A Primer

International Journals of Advanced Research in Computer Science and Software Engineering ISSN: 2277-128X (Volume-8, Issue-5) a Research Article May 2018 Swarm Intelligence: A Primer Matthew N. O. Sadiku, Mahamadou Tembely, and Sarhan M. Musa Roy G. Perry College of Engineering, Prairie View A&M University, Prairie View, TX 77446, United States Email: sadiku@ieee.org; mtembely@student.pvamu.edu; smmusa @pvamu.edu Abstract— Swarm intelligence is the emergent collective intelligence of groups of simple agents. It belongs to the emerging field of bio-inspired soft computing. It is inspired from the biological entities such as birds, fish, ants, wasps, termites, and bees. Bio-inspired computation is a field of study that is closely related to artificial intelligence. This paper provides a brief introduction to swarm intelligence. Keywords— swarm intelligence, intelligent behavior, artificial intelligence I. INTRODUCTION A swarm may be regarded as group of individuals who have chosen to converge on a common goal. Swarms consist of a large number of simple entities that interact with each other and with the environment with no central control. The concept of swarm intelligence is an emerging domain which belongs to the field of artificial intelligence (AI), as many of its pursuits can be linked to machine learning. Swarm intelligence (SI) is nature-inspired, especially biological systems [1]. Recent studies have shown that individual social insects such as ants or bees follow simple rules and have no centralized control. The collective behavior of the swarm presents an intelligence that is beyond the intelligence of each individual member. In other words, a group of bees works better than one alone. In the bee colony, each bee has its own division of labor and unique work. A human swarm is a social network of individuals behaving for a time like-mindedly. The more sophisticated the brain, the more sophisticated the swarm behavior [2]. Some researchers even consider the collective, the swarm, as one single and unique individual. . II. SI SYSTEMS AND TECHNIQUES Swarm Intelligence was introduced by Gerardo Beni and Jing Wang in 1989 with their study of cellular robotic systems [3]. The scientific discipline was born from biological insights about the incredible abilities of social insects to solve their everyday-life problems. SI systems consist typically of a group of simple agents interacting with each other and with their environment. Typical examples of natural SI systems include ant colonies, bird flocking, and animal herding. Swarm intelligence-based techniques are recent computational tools for systems optimization. They are population-based stochastic methods used in combinatorial optimization problems. The two most popular of them are: particle (individual) swarm optimization (PSO) and ant colony optimization (ACO). PSO is a population-based algorithm for handling problems in which a point best represents a solution. It searches for an optimal solution in the computable search space. It is inspired by the intelligent, experience-sharing, social behavior of birds flocking to find the best food area, schools of fish, swarms of bees, or even human social behavior. ACO is a metaheuristic for solving combinatorial optimization problems. It draws inspiration from the behavior of ant colonies. While exploring their environment, real ants lay down pheromones directing each other to where resources are located. Besides these two popular SI algorithms, others include bee-inspired algorithms, bacterial foraging optimization, firefly algorithms, artificial bee colony, and fish swarm optimization. These methods have been proposed to train artificial neural networks and other real-world problems in science and engineering [4,5]. These SI techniques may be regarded as another set of soft computing techniques. III. APPLICATIONS Swarm intelligence-based algorithms have been applied in almost all areas of sciences and engineering. SI is used for solving problems in a variety of areas of application such as robotics, engineering, education, MANET, and forecasting.  Robots: The application of swarm techniques to robots is called swarm robotics. Swarm robotics may be regarded as the study of how a swarm of simple agents can be constructed to collectively accomplish tasks that © www.ijarcsse.com, All Rights Reserved Page | 100 Sadiku et al., International Journal of Advanced Research in Computer Science and Software Engineering 8(5) ISSN(E): 2277-128X, ISSN(P): 2277-6451, pp. 100-102 are beyond the capability of a single one. It is a novel approach for coordinating a large number of robots [6]. A robot swarm is an autonomous entity that acts in a self-organized manner. It does not have centralized entity that is responsible for its coordination. Designing the robot–to-robot or robot-to-environment interactions that would result in the desired collective behavior is very challenging. It is possible to of use swarm intelligence to control nanobots within the body for the purpose of killing cancer tumors. SI is well suited for multi-robot motion planning [7].  Engineering: SI techniques are useful tools for solving practical engineering problems. SI algorithms have been applied successfully to solve optimization problems in mechanical engineering [8].  Education: In the educational system, swarm intelligence has been applied to curriculum planning, computer adaptive testing, intelligent assessment paper generation, and class/examinations scheduling [9]. Swarm intelligence can be adopted into a library system so that the system becomes intelligent. As long as the readers can access the Internet, they can conveniently obtain book information in the library that is closest to them [10].  MANET: A mobile ad hoc network (MANET) is a self-organizing multihop wireless network where the structure of network changes dynamically. Routing in MANET is very difficult due to its dynamic topology and limited bandwidth. A new set of routing protocols based on swarm intelligence for MANETs has emerged. They are more capable of handling various routing problems such as scalability, maintainability, energy utilization, and high mobility [11, 12]. In addition, SI can be used to solve the complex combinatorial optimization problems, non-linear design system optimization, and biometric features selection and optimization, stock market price movement, face recognition, meeting scheduling, water resources management, routing in telecommunication networks, and data mining [13]. IV. BENEFITS AND CHALLENGES SI systems have the following advantages [14]: scalability, adaptability, flexibility, collective robustness, and individual simplicity. We can exploit the social collective behavior of swarms to solve complex real-life problems. A potential disadvantage to swarm intelligence compared to decisions being made by a minority of individuals is its speed. Decisions where speed is a critical factor may show a simpler network of inter-individual interactions to minimise the number of connections and maximize response speed [15]. Engineering design of systems based on the swarm intelligence paradigm can be challenging. A lot of work needs to be done before dependable robotic swarms can become an engineering reality. V. CONCLUSION Swarm intelligence, inspired by social insects, has emerged in recent years as an innovative AI technique for solving complex problems. The concept refers to a kind of problem-solving ability that emerges in the interactions of simple information-processing units. The information-processing units that compose a swarm can be animate, insects, birds, human beings, robots, or standalone workstations [7]. The backbone of SI is built on two families of algorithms: ant colony optimization, and particle swarm optimization. For more information, one should consult [16-20] and other books in Amazon.com. One should also consult the three journals exclusively devoted to sward intelligence: International Journal of Swarm Intelligence Research, Swarm and Evolutionary Computation (Elsevier), and Swarm Intelligence (Springer). REFERENCES [1] Swarm intelligence,” Wikipedia, the free encyclopedia, https://en.wikipedia.org/wiki/Swarm_intelligence [2] J. H. Rolling, “Swarm intelligence and collaboration,” Art Education, vol. 69, no. 5, 2016, pp. 4-6. [3] G. Beni and J. Wang ., “Swarm intelligence,” Proceedings of the 7th Annual Meeting of the Robotics Society of Japan,1989, pp. 425-428. [4] M. Conforth and Y. Meng, “Reinforcement learning using swarm intelligence-trained neural networks,” Journal of Experimental & Theoretical Artificial Intelligence, vol. 22, no. 3, 2010, pp. 197-218. [5] M. J. R. Mish, “Swarm Intelligence techniques and its applications in water resources management,” ISH Journal of Hydraulic Engineering, vol. 15, sup1, 2009, pp. 151-169. [6] E. Sahin and W. M. Spears (eds.), Swarm Robotics. New York: Springer, 2005. [7] G. G. Rigatos, “Multi-robot motion planning using swarm intelligence,” International Journal of Advanced Robotic Systems, vol. 5, no. 2, 2008, pp. 139-144. © www.ijarcsse.com, All Rights Reserved Page | 101 [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] Sadiku et al., International Journal of Advanced Research in Computer Science and Software Engineering 8(5) ISSN(E): 2277-128X, ISSN(P): 2277-6451, pp. 100-102 B. Yu and G. Chen, “Swarm intelligence in mechanical engineering,” Advances in Mechanical Engineering, 2016, vol. 8, no. 12, 2016, pp. 1–3. L. H. Wong and C. K. Looi, “Swarm intelligence: new techniques for adaptive systems to provide learning support,” Interactive Learning Environments, vol. 20, no. 1, 2012, pp. 19-40. L. S. Chen, ”Applying swarm intelligence to a library system,” Library Collections, Acquisitions, and Technical Services, vol. 34, no. 1, 2010, pp. 1-10. C. C. Ioannou, “Swarm intelligence in fish? The difficulty in demonstrating distributed and self-organised collective intelligence in (some) animal groups,” Behavioural Processes, vol.141, 2017, pp. 141–151. N. Srivastava and P. Raghav, “A review on swarm intelligence based routing algorithms in mobile adhoc network,” Proceeding of the 8th International Conference on Computing, Communication and Networking Technologies, Delhi, India, 2017. A. Giagkos and M. S. Wilson, “Swarm intelligence to wireless ad hoc networks: adaptive honeybee foraging during communication sessions,” Adaptive Behavior, vol. 21, no. 6, 2013, pp. 501–515. A. Saggu1, P. Yadav, and M. Roopak, “Applications of swarm intelligence,” International Journal of Computer Science and Mobile Computing, vol. .2, no.. 5, May 2013, pp. 353-359. H. Ahmed and J. Glasgow, “Swarm intelligence: Concepts, models and applications,” https://pdfs.semanticscholar.org/116b/67cf2ad2c948533e6890a9fccc5543dded89.pdf J. Kennedy. “Smarm Intelligence,” in A. Y. Zomaya (ed.), Handbook of Nature-Inspired and Innovation Computing. New York: Springer, chapter 6, 2006, pp. 187-219. L. P. Walters, Applications of Swarm Intelligence. Nova Science Publishers, 2010. R. Eberhart,Y. Shi, and J. Kennedy, Swarm Intelligence. Morgan Kaufmann, 2001 A. P. Engelbrecht, Fundamentals of Computational Swarm Intelligence. John Wiley & Sons, 2006. C. Blum and D. Merkle (eds.), Swarm Intelligence: Introduction and Applications. Springer, 2008. Y. Tan, Y, Shi, and T. K. Chen (eds), Advances in Swarm Intelligence. Berlin, Germany: Springer-Verlag, 2010. ABOUT THE AUTHORS Matthew N.O. Sadiku is a professor in the Department of Electrical and Computer Engineering at Prairie View A&M University, Prairie View, Texas. He is the author of several books and papers. His areas of research interest include computational electromagnetics and computer networks. He is a fellow of IEEE. Mahamadou Tembely is an adjunct professor Ph.D at Prairie View A&M University, Texas. He received the 2014 Outstanding MS Graduated Student award for the department of electrical and computer engineering. He is the author of several papers. Sarhan M. Musa is a professor in the Department of Engineering Technology at Prairie View A&M University, Texas. He has been the director of Prairie View Networking Academy, Texas, since 2004. He is an LTD Sprint and Boeing Welliver Fellow. © www.ijarcsse.com, All Rights Reserved Page | 102