Showing papers on "Swarm intelligence published in 2002"

Particle swarm optimization

Abstract: A concept for the optimization of nonlinear functions using particle swarm methodology is introduced. The evolution of several paradigms is outlined, and an implementation of one of the paradigms is discussed. Benchmark testing of the paradigm is described, and applications, including nonlinear function optimization and neural network training, are proposed. The relationships between particle swarm optimization and both artificial life and genetic algorithms are described.

Swarm Intelligence: From Natural to Artificial Systems

Abstract: (2002). Swarm Intelligence: From Natural to Artificial Systems. Connection Science: Vol. 14, No. 2, pp. 163-164.

Recent approaches to global optimization problems through Particle Swarm Optimization

Abstract: This paper presents an overview of our most recent results concerning the Particle Swarm Optimization (PSO) method. Techniques for the alleviation of local minima, and for detecting multiple minimizers are described. Moreover, results on the ability of the PSO in tackling Multiobjective, Minimax, Integer Programming and e1 errors-in-variables problems, as well as problems in noisy and continuously changing environments, are reported. Finally, a Composite PSO, in which the heuristic parameters of PSO are controlled by a Differential Evolution algorithm during the optimization, is described, and results for many well-known and widely used test functions are given.

Adaptive particle swarm optimization on individual level

Abstract: An adaptive particle swarm optimization (PSO) on individual level is presented. By analyzing the social model of PSO, a replacement criterion, based on the diversity of fitness between the current particle and the best historical experience, is introduced to maintain the social attribution of swarm adaptively by taking off inactive particles. The testing of three benchmark functions indicates that it improves the average performance effectively.

Transportation modeling: an artificial life approach

Abstract: Artificial life (ALife) uses biological knowledge and techniques to help solve different engineering, management, control and computational problems. Natural systems teach us that very simple individual organisms can form systems capable of performing highly complex tasks by dynamically interacting with each other. The main goal of this paper is to show how we can use ALife concepts (inspired by some principles of natural swarm intelligence) when solving complex problems in traffic and transportation. The bee system that represents the new approach in the field of swarm intelligence is described. It is also shown in the paper that ALife approach can be successful to "attack" transportation problems characterized by uncertainty. The fuzzy ant system (FAS) described in the paper represents an attempt to handle the uncertainty that sometimes exists in some complex transportation problems. The potential applications of the bee system and the fuzzy ant system in the field of traffic and transportation engineering are discussed.

Tracking changing extrema with adaptive particle swarm optimizer

Abstract: A modification of the particle swarm optimizer (PSO) involving updating obsolete particle memories has been shown to be effective in locating a changing extrema. In this paper we investigate the effectiveness of the modified PSO in tracking changing extrema over time. We demonstrate that a modified PSO, called the adaptive PSO, reliably and accurately tracks a continuously changing solution.

Swarm intelligence: power in numbers

Abstract: Following a trail of insects as they work together to accomplish a task offers unique possibilities for problem solving.

Self-organized flocking with agent failure: Off-line optimization and demonstration with real robots

Abstract: This paper presents an investigation of flocking by teams of autonomous mobile robots using principles of Swarm Intelligence. First, we present a simple flocking task, and we describe a leaderless distributed flocking algorithm (LD) that is more conducive to implementation on embodied agents than the established algorithms used in computer animation. Next, we use an embodied simulator and reinforcement learning techniques to optimize LD performance under different conditions, showing that this method can be used not only to improve performance but also to gain insight into which algorithm components contribute most to system behavior. Finally, we demonstrate that a group of real robots executing LD with emulated sensors can successfully flock (even in the presence of individual agent failure) and that systematic characterization (and therefore optimization) of real robot flocking performance is achievable.

Improvised music with swarms

Abstract: This paper describes SWARMUSIC, an interactive music improviser. A particle swarm algorithm generates musical material by a mapping of particle positions onto events in MIDI space. Interaction with an external musical source arises through the attraction of the particle swarm to a target. SWARMUSIC is the first application of swarm intelligence to music.

SWARM-BOT: pattern formation in a swarm of self-assembling mobile robots

Abstract: We introduce a new robotic system, called swarm-bot. The system consists of a swarm of mobile robots with the ability to connect to/disconnect from each other to self-assemble into different kinds of structures. First, we describe our vision and the goals of the project. Then we present preliminary results on the formation of patterns obtained from a grid-world simulation of the system.

A Robotic Swarm for Spill Finding and Perimeter Formation

Abstract: : This paper addresses issues surrounding deployment and tasking of a real-world collective of cost-effective, small mobile robots. To escape the limitations of centralized control, this project distributes control using an innovative, multi-modal communication architecture including acoustical chirping, infrared, and radio frequency transmissions. This paper reports on the use of social potential fields - attractive and repulsive fields emitted by each robot -- as a means to coordinate group behavior and promote the emergence of swarm intelligence as seen in a colony of ants or swarm of bees. A suite of C2 tools, AgentTools, has been developed to enable an operator to inject high-level domain knowledge and guidance into the behavior of the otherwise autonomous robots. The resulting system permits the user to interact with functional groups, rather than issuing commands to each individual robot. Using the realworld robot collective and C2 system, the Idaho National Engineering and Environmental Laboratory has performed experiments to empirically analyze the benefits and limitations associated with the use of many small-scale robots. Experimental results point to fundamental advantages of distributed systems and indicate that our real-world implementation of social potential fields scales well to varying numbers of robots and improves performance in terms of time and reliability.

Co-fields: towards a unifying approach to the engineering of swarm intelligent systems

Abstract: Swarm intelligent systems, in which the paths to problem solving emerge as the result of interactions between simple autonomous components (agents or ants) and between them and their environment, appear very promising to develop robust and flexible software application. However, the variety of swarm-based approaches that have been proposed so far still lacks a common modeling and engineering methodology. In the attempt to overcome this problem, this paper presents a general coordination methodology in which swarm's components are simply driven by abstract computational force fields (Co-Fields), generated either by agents, or by the environment. By having agents be driven in their activities by such fields, globally coordinated behaviors can naturally emerge. Although this model still does not offer a complete engineering methodology, it can provide a unifying abstraction for swarm intelligent systems and it can also be exploited to formalize these systems in terms of dynamical systems whose behavior can be described via differential equations. Several example of swarm systems modeled with Co-Fields are presented to support our thesis.

CSIM: a document clustering algorithm based on swarm intelligence

Abstract: This paper presents a document clustering algorithm based on swarm intelligence and k-means: CSIM. First, a document clustering algorithm based on swarm intelligence is employed. It is derived from a basic model interpreting ant colony organization of cemeteries. Swarm intelligence for flexibility, self-organization and robustness has been applied in a variety of areas. Taking advantage of these traits, good initial clusters are obtained in the first phase in CSIM. We then combine it with the classical k-means clustering method by using the clusters as initial centers. CSIM inherits the prominent properties of both swarm intelligence and k-means. It also offsets the weakness of those two techniques. Experimental results show the good performance of the hybrid document clustering algorithm.

Modeling transportation problems using concepts of swarm intelligence and soft computing

Abstract: Many real-world problems could be formulated in a way to fit the necessary form for discrete optimization. Discrete optimization problems could be solved by numerous different techniques that have appeared through years. Some of the techniques will provide optimal solution(s) to the problem and some of them will give “good enough” solution(s). Fundamental reason for developing techniques capable of producing solutions that are not necessarily optimal is the fact that many of discrete optimization problems are NP-complete. Metaheuristic algorithms are a common name for a set of general purpose techniques developed to provide solution to the problems belonging to discrete optimization. Mostly the techniques are based on natural metaphors. Countless problems in transportation engineering could be formulated as discrete optimization problems. Recently, researchers started studying the behavior of social insects (ants) in an attempt to use the swarm intelligence concept to develop artificial systems with the ability to search a problem's solution space in a way that is similar to the foraging search by a colony of social insects. The development of artificial systems does not entail the complete imitation of natural systems, but explores them in search of ideas for modeling. This research is partially devoted to the development of a new system based on foraging behavior of bee colonies - Bee System. The Bee System was tested through many instances of the Traveling Salesman Problem. Many transportation-engineering problems besides being of combinatorial nature are characterized by uncertainty. In order to treat these problems, the second part of the research is devoted to development of the algorithms combining existing results in the area of swarm intelligence (existing Ant System) and approximate reasoning. The proposed approach—Fuzzy Ant System is tested on the following two examples: Stochastic Vehicle Routing Problem and Schedule Synchronization in Public Transit.

Adaptive-SDR: adaptive swarm-based distributed routing

Abstract: Swarm intelligence forms the core of a new class of algorithms inspired by the social behavior of insects that live in swarms. Its attractive features include adaptation, robustness and a distributed, decentralized nature, rendering swarm-based algorithms well-suited for routing in wireless or satellite networks, where it is difficult it implement centralized network control. We propose one such routing algorithm, dubbed adaptive swarm-based distributed routing (adaptive-SDR), which is scalable, robust and suitable to handle large amounts of network traffic, while minimizing delay and packet loss.

Swarm intelligence on the binary constraint satisfaction problem

Abstract: We introduce a discrete particle swarm (PS) algorithm for solving binary constraint satisfaction problems (CSPs). It uses information about the conflicts between the variables to compute the velocity of the individual particles. We tune the parameters of the PS algorithm to a quasi-optimal setting and study the behavior of the algorithm under changes to this setting. The PS algorithm is then empirically compared with ant colonies (which also belong to the swarm intelligence class) and genetic algorithms on a whole range of randomly generated binary CSP instances.

An Adaptive Flocking Algorithm for Spatial Clustering

Abstract: This paper presents a parallel spatial clustering algorithm based on the use of new Swarm Intelligence (SI) techniques. SI is an emerging new area of research into Artificial Life, where a problem can be solved using a set of biologically inspired (unintelligent) agents exhibiting a collective intelligent behaviour. The algorithm, called SPARROW, combines a smart exploratory strategy based on a flock of birds with a density-based cluster algorithm to discover clusters of arbitrary shape and size in spatial data. Agents use modified rules of the standard flock algorithm to transform an agent into a hunter foraging for clusters in spatial data. We have applied this algorithm to two synthetic data sets and we have measured, through computer simulation, the impact of the flocking search strategy on performance. Moreover, we have evaluated the accuracy of SPARROW compared to the DBSCAN algorithm.

Immune, swarm, and evolutionary algorithms. Part II: philosophical comparisons

Abstract: For Part I see ibid. vol. 3 (2002). In the first part of this paper, the standard evolutionary, immune, and swarm algorithms were reviewed. This second part starts by presenting a philosophical discussion about some similarities and differences among the various approaches in terms of their basic components, structure, knowledge storage, adaptation paradigm, interactions, and metaphor. Then, the identification of the main features of each technique is performed in order to shed some light into how to create hybrid algorithms.

SWARM-BOT: A Swarm of Autonomous Mobile Robots with Self-Assembling Capabilities

Abstract: We present a new robotic concept, called SWARM-BOT, based on a swarm of small and simple autonomous mobile robots called S-BOTs. S-BOTs have a particular assembling capability that allows them to connect physically to other S-BOTs and form a bigger robot entity, the SWARM-BOT. A SWARM-BOT is typically composed by 10 to 30 S-BOTs physically interconnected. S-BOTs can autonomously assemble into a SWARM-BOT but also disassemble again. This feature of the S-BOTs provides SWARM-BOT with self-assembling and self-reconfiguring capabilities. Such a concept, by taking advantage from the collective and distributed approaches, ensures robustness to failures even in hard environment conditions. The approach presented finds its theoretical roots in recent studies on swarm intelligence.

A swarm intelligence based approach to the mine detection problem

Abstract: In this paper, we have applied a swarm intelligence based technique to a mine detection problem. Swarm intelligence techniques are used to model robotic agents to solve the problem. Studies made on the ant colonies, which is a typical member of the family of swarms, are applied in devising the techniques for the agents. Ant colony models bestow intelligence not only at the individual level, but more at the collective level (the interactions produced by the individual members in trying to solve a common problem). An analysis of the results obtained with a computer simulation of the mine detection is also presented.

Cooperative Behavior of Interacting Simple Robots in a Clockface Arranged Foraging Field

Abstract: Emergence of swarm intelligence is investigated through the foraging task. As the robots assumed in this paper is very simple in order to discuss their behavior analytically, there are a few parameters to characterize their behavior such as interaction duration and interaction range. The behavior of the group is investigated on computer simulation and mathematical model. In this paper, we discuss the definition of swarm intelligence at first, and shows the emergence of swarm intelligence in a clockface arranged foraging field.

Sensor Communication Network Using Swarm Intelligence

Abstract: Energy consumption is currently a key issue in research for future sensor networks. This paper presents a novel approach to sensor network routing based on energy consumption. The unique routing algorithm uses swarm intelligence, which is computationally efficient.

Modeling robot swarms using agent-based simulation

Abstract: : In the near future advances in mechanical and electrical engineering will enable the production of a wide variety of relatively low cost robotic vehicles. This thesis investigates the behavior of swans of military robots acting autonomously. The Multi-Agent Robot Swarm Simulation (MARSS) was developed for modeling the behavior of swarm of military robots. MARSS contains state, sensing, and behavioral model building tools that allow a range of complex entities and interactions to be represented. It is a model-building tool that draws theory and ideas from agent-based simulation, discrete event simulation, traditional operations research, search theory, swarm theory, and experimental design. MARS S enables analysts to explore the effect of individual behavioral factors on swarm performance. The performance response surface can be explored using designed experiments. A model was developed in MARS S to investigate the effects of increasing behavioral complexity for a search scenario involving a swarm of Micro Air Vehicles (MAV's) searching for mobile tanks in a region. Agreement between theoretical and simulated search scenarios for simple searchers was found. The effect of increased MAV sensory and behavioral capability was demonstrated to be important. Little improvement was observed in swarm performance with these capabilities, however agent performance was adversely affected by reacting to increased knowledge in the wrong way. The utility of MARSS for conducting this type of analysis was demonstrated.

Classifying English Web pages with "smart" ant-like agents

Abstract: As the amount of online information in the form of web pages grows, the demand of text categorization to assist in efficient retrieval will surely increase in tandem. Even though such a task may be performed manually by the domain experts, it is unlikely that such human categorization will be able to keep pace with the tremendous growth of the World Wide Web (WWW). The sheer volume of data that is available on the Internet will surely overwhelm these human categorizers. Hence, as the Internet expands, the importance of having this process automated will become increasingly important. Many clustering techniques have been commonly used to retrieve, filter, and categorize documents. The traditional clustering algorithms either use a priori knowledge of document structures to define a distance or similarity among these documents, or use probabilistic techniques such as Bayesian classification. In recent studies, several species of ants have been observed to exhibit collective behavior that has inspired the computation algorithm commonly known as swarm intelligence. This has been used in a variety of different applications, ranging from exploratory data analysis, scheduling, to graph partitioning. This paper investigates the use of an ant-based algorithm to classify English Web.

Implementation of Ant Colony Optimization Algorithm for Mobile Ad hoc Network Applications: OpenMP Experiences

Abstract: There is a large class of applications where the pattern of data distributions is non-uniform and sparse. The algorithms for these applications are usually asynchronous. These applications are generally classified as irregular problems. From the parallel computing perspective, an ad hoc network is one such application since the network changes dynamically at runtime, exhibits chaotic load balancing and unpredictable communication behavior among the nodes during runtime. Ant Colony Optimization (ACO) meta-heuristic, a subset of swarm intelligence, is an inherently parallelizable search technique recently proposed to determine routing in ad hoc networks. One of the many interesting features of swarm based approach is their ability to solve problems that are not static but are spatially distributed and changing over time. In this paper we report our experiences in design, development and implementation of a parallel algorithm for mobile ad hoc networks (MANETs) using the ACO technique on a shared memory architecture with OpenMP. We have experimented with three scheduling policies provided by OpenMP for varying data sizes. Also we report comparison of the performance results with message passing environment (MPI).

Simulating evolutionary agent communities with OOCSMP

Abstract: This paper describes some extensions added to the continuous simulation language OOCSMP to perform agent-oriented simulation. The extensions are tested by simulating the evolution of a colony of virtual ants (vants). In this simulation, each vant is modelled as an agent and is assigned a set of genes that control some aspects of its behaviour, such as its velocity, memory, communication abilities, scepticism, etc. Some emergent properties of the swarm of vants have been observed.