Showing papers in "Swarm Intelligence in 2008"

Swarm Intelligence in Optimization

Abstract: Optimization techniques inspired by swarm intelligence have become increasingly popular during the last decade. They are characterized by a decentralized way of working that mimics the behavior of swarms of social insects, flocks of birds, or schools of fish. The advantage of these approaches over traditional techniques is their robustness and flexibility. These properties make swarm intelligence a successful design paradigm for algorithms that deal with increasingly complex problems. In this chapter we focus on two of the most successful examples of optimization techniques inspired by swarm intelligence: ant colony optimization and particle swarm optimization. Ant colony optimization was introduced as a technique for combinatorial optimization in the early 1990s. The inspiring source of ant colony optimization is the foraging behavior of real ant colonies. In addition, particle swarm optimization was introduced for continuous optimization in the mid-1990s, inspired by bird flocking.

Massively multi-robot simulation in stage

Abstract: Stage is a C++ software library that simulates multiple mobile robots. Stage version 2, as the simulation backend for the Player/Stage system, may be the most commonly used robot simulator in research and university teaching today. Development of Stage version 3 has focused on improving scalability, usability, and portability. This paper examines Stage’s scalability.

Self-organized flocking in mobile robot swarms

Abstract: In this paper, we study self-organized flocking in a swarm of mobile robots. We present Kobot, a mobile robot platform developed specifically for swarm robotic studies. We describe its infrared-based short range sensing system, capable of measuring the distance from obstacles and detecting kin robots, and a novel sensing system called the virtual heading system (VHS) which uses a digital compass and a wireless communication module for sensing the relative headings of neighboring robots.

Path formation in a robot swarm

Abstract: We present two swarm intelligence control mechanisms used for distributed robot path formation. In the first, the robots form linear chains. We study three variants of robot chains, which vary in the degree of motion allowed to the chain structure. The second mechanism is called vectorfield. In this case, the robots form a pattern that globally indicates the direction towards a goal or home location.

A framework of space–time continuous models for algorithm design in swarm robotics

Abstract: Designing and analyzing self-organizing systems such as robotic swarms is a challenging task even though we have complete knowledge about the robot’s interior. It is difficult to determine the individual robot’s behavior based on the swarm behavior and vice versa due to the high number of agent–agent interactions. A step towards a solution of this problem is the development of appropriate models which accurately predict the swarm behavior based on a specified control algorithm. Such models would reduce the necessary number of time-consuming simulations and experiments during the design process of an algorithm. In this paper we propose a model with focus on an explicit representation of space because the effectiveness of many swarm robotic scenarios depends on spatial inhomogeneity. We use methods of statistical physics to address spatiality. Starting from a description of a single robot we derive an abstract model of swarm motion. The model is then extended to a generic model framework of communicating robots. In two examples we validate models against simulation results. Our experience shows that qualitative correctness is easily achieved, while quantitative correctness is disproportionately more difficult but still possible.

Particle Swarms for Dynamic Optimization Problems

Abstract: Many practical optimization problems are dynamic in the sense that the best solution changes in time. An optimization algorithm, therefore, has to both find and subsequently track the changing optimum. examples include the arrival of new jobs in scheduling, changing expected profits in portfolio optimization, and fluctuating demand. Clearly, if the changes in the problem instance are radical, the best one can do is to repeatedly solve the optimization problem from scratch. However, in most practical applications the changes are gradual. If this is the case, it should be possible to speed up optimization after a problem change by utilizing some of the information on the fitness landscape gathered during the optimization process so far. In recent years, appropriately modified evolutionary algorithms (EAs) have been shown to achieve this successfully; see, e.g., [11, 23]; the focus of this chapter is to present similar advances within the context of particle swarm optimization. Particle swarm optimization (PSO) is a versatile population-based optimization technique, in many respects similar to evolutionary algorithms. Basically, particles 'fly' above the fitness landscape, while a particle's movement is influenced by its attraction to its neighborhood best (the best solution found by members of the particle's social network), and its personal best (the best solution the particle has found so far). PSO has been shown to perform well for many static problems [32], and is introduced in more detail in Section 2.

Biologically inspired redistribution of a swarm of robots among multiple sites

Abstract: We present a biologically inspired approach to the dynamic assignment and reassignment of a homogeneous swarm of robots to multiple locations, which is relevant to applications like search and rescue, environmental monitoring, and task allocation. Our work is inspired by experimental studies of ant house hunting and empirical models that predict the behavior of the colony that is faced with a choice between multiple candidate nests. We design quorum based stochastic control policies that enable the team of agents to distribute themselves among multiple candidate sites in a specified ratio, and compare our results to the linear stochastic policies described in (Halasz et al., in Proceedings of the International Conference on Intelligent Robots and Systems (IROS’07), pp. 2320–2325, 2007). We show how our quorum model consistently performs better than the linear models while minimizing computational requirements and now it can be implemented without the use of inter-agent wireless communication.

Special Issue on Swarm Robotics

Abstract: Swarm robotics is a new approach to the coordination of multi-robot systems. In contrast with traditional multi-robot systems which use centralised or hierarchical control and communication systems in order to coordinate robots’ behaviours, swarm robotics adopts a decentralised approach in which the desired collective behaviours emerge from the local interactions between robots and their environment. Such emergent or self-organised collective behaviours are inspired by, and in some cases modelled on, the swarm intelligence observed in social insects. The potential for swarm robotics is considerable. Any task in which physically distributed objects need to be explored, surveyed, collected, harvested, rescued, or assembled into structures is a potential real-world application for swarm robotics. The key advantage of the swarm robotics approach is robustness, which manifests itself in a number of ways. Firstly, because a swarm of robots consists of a number of relatively simple and typically homogeneous robots, which are not pre-assigned to specific roles or tasks within the swarm, then the swarm can self-organise or dynamically re-organise the way individual robots are deployed. Secondly, and for the same reasons, the swarm approach is highly tolerant to the failure of individual robots. Thirdly, the fact that control is completely decentralised means that there is no common-mode failure point or vulnerability in the swarm. Indeed, it could be said that the high level of robustness evident in robotic swarms comes for free in the sense that it is intrinsic to the swarm robotics approach, which contrasts with the high engineering cost of fault tolerance in conventional robotic systems. The realisation of the potential of swarm robotics requires the solution of a number of very challenging problems. Firstly, in algorithm design: swarm roboticists face the problem of designing both the physical morphology and behaviours of the individual robots such that when those robots interact with each other and their environment, the desired overall collective behaviours will emerge. At present there are no principled approaches to the design of low-level behaviours for a given desired collective behaviour. Secondly, in implementation

Modelling a wireless connected swarm of mobile robots

Abstract: It is a characteristic of swarm robotics that modelling the overall swarm behaviour in terms of the low-level behaviours of individual robots is very difficult. Yet if swarm robotics is to make the transition from the laboratory to real-world engineering realisation such models would be critical for both overall validation of algorithm correctness and detailed parameter optimisation. We seek models with predictive power: models that allow us to determine the effect of modifying parameters in individual robots on the overall swarm behaviour. This paper presents results from a study to apply the probabilistic modelling approach to a class of wireless connected swarms operating in unbounded environments. The paper proposes a probabilistic finite state machine (PFSM) that describes the network connectivity and overall macroscopic behaviour of the swarm, then develops a novel robot-centric approach to the estimation of the state transition probabilities within the PFSM. Using measured data from simulation the paper then carefully validates the PFSM model step by step, allowing us to assess the accuracy and hence the utility of the model.

Routing Protocols for Next Generation Networks Inspired by Collective Behaviors of Insect Societies: An Overview

Abstract: In this chapter we discuss the properties and review the main instances of network routing algorithms whose bottom-up design has been inspired by collective behaviors of social insects such as ants and bees. This class of bio-inspired routing algorithms includes a relatively large number of algorithms mostly developed during the last ten years. The characteristics inherited by the biological systems of inspiration almost naturally empower these algorithms with characteristics such as autonomy, self-organization, adaptivity, robustness, and scalability, which are all desirable if not necessary properties to deal with the challenges of current and next-generation networks. In the chapter we consider different classes of wired and wireless networks, and for each class we briefly discuss the characteristics of the main ant- and bee-colony-inspired algorithms which can be found in literature. We point out their distinctive features and discuss their general pros and cons in relationship to the state of the art.

Ant-based swarming with positionless micro air vehicles for communication relay

Abstract: Swarming without positioning information is interesting in application- oriented systems because it alleviates the need for sensors which are dependent on the environment, expensive in terms of energy, cost, size and weight, or unusable at useful ranges for real-life scenarios. This principle is applied to the development of a swarm of micro air vehicles (SMAVs) for the deployment of ad hoc wireless communication networks (SMAVNETs) between ground users in disaster areas. Rather than relying on positioning information, MAVs rely on local communication with immediate neighbors and proprioceptive sensors which provide heading, speed and altitude. To solve the challenging task of designing agent controllers to achieve the swarm behavior of the SMAVNET, inspiration is taken from army ants which are capable of laying and maintaining pheromone paths leading from their nest to food sources in nature. This is analogous to the deployment of communication pathways between multiple ground users. However, instead of being physically deposited in the air or on a map, pheromone is virtually deposited on the MAVs using local communication. This approach is investigated in 3D simulation in a simplified scenario with two ground users.

Evolving coordinated group behaviours through maximisation of mean mutual information

Abstract: A well known problem in the design of the control system for a swarm of robots concerns the definition of suitable individual rules that result in the desired coordinated behaviour. A possible solution to this problem is given by the automatic synthesis of the individual controllers through evolutionary or learning processes. These processes offer the possibility to freely search the space of the possible solutions for a given task, under the guidance of a user-defined utility function. Nonetheless, there exist no general principles to follow in the definition of such a utility function in order to reward coordinated group behaviours. As a consequence, task dependent functions must be devised each time a new coordination problem is under study. In this paper, we propose the use of measures developed in Information Theory as task-independent, implicit utility functions. We present two experiments in which three robots are trained to produce generic coordinated behaviours. Each robot is provided with rich sensory and motor apparatus, which can be exploited to explore the environment and to communicate with other robots. We show how coordinated behaviours can be synthesised through a simple evolutionary process. The only criteria used to evaluate the performance of the robotic group is the estimate of mutual information between the motor states of the robots.

Biological Foundations of Swarm Intelligence

Abstract: Why should a book on swarm intelligence start with a chapter on biology? Because swarm intelligence is biology. For millions of years many biological systems have solved complex problems by sharing information with group members. By carefully studying the underlying individual behaviours and combining behavioral observations with mathematical or simulation modeling we are now able to understand the underlying mechanisms of collective behavior in biological systems. We use examples from the insect world to illustrate how patterns are formed, how collective decisions are made and how groups comprised of large numbers of insects are able to move as one. We hope that this first chapter will encourage and inspire computer scientists to look more closely at biological systems.

SWARMORPH-script: a language for arbitrary morphology generation in self-assembling robots

Abstract: In certain multi-robot systems, the physical limitations of the individual robots can be overcome using self-assembly—the autonomous creation of physical connections between individual robots to form a larger composite robotic entity. However, existing robotic systems capable of self-assembly have little or no control over the morphology of the self-assembled entities. This restricts the adaptability of such systems, since robots can carry out certain tasks more efficiently if their morphology is specialized to the task. In this paper, we extend the distributed mechanism presented in (Christensen et al. in IEEE Robot. Autom. Mag. 14(4):18–25, 2007) that allows autonomous mobile robots to self-assemble into specific morphologies. We present a simple language, SWARMORPH-script, that allows for concise descriptions of the rules that govern the distributed morphology growth process. Local visual communication allows physically connected robots to send and receive strings. A string can be a rule identifier that triggers execution of predefined logic for extending a morphology. Alternatively, whole scripts can be communicated and subsequently executed on the receiving robot. On real self-propelled robots capable of self-assembly, we demonstrate how specific morphologies can be constructed, how the size of a morphology can be regulated, and how multiple morphologies can be assembled. We also show how the transmission of entire scripts gives the robots the capacity to participate in the formation of morphologies of which they had no a priori knowledge.

Evolution, Self-organization and Swarm Robotics

Abstract: The activities of social insects are often based on a self-organising process, that is, “a process in which pattern at the global level of a system emerges solely from numerous interactions among the lower-level components of the system”(see Camazine-EtAl:01, p. 8). In a self-organising system such as an ant colony, there is neither a leader that drives the activities of the group, nor are the individual ants informed about a global recipe or blueprint to be executed. On the contrary, each single ant acts autonomously following simple rules and locally interacting with the other ants. As a consequence of the numerous interactions among individuals, a coherent behaviour can be observed at the colony level.

Characteristics of ant-inspired traffic flow

Abstract: We investigate the organization of traffic flow on preexisting uni- and bidirectional ant trails. Our investigations comprise a theoretical as well as an empirical part. We propose minimal models of uni- and bi-directional traffic flow implemented as cellular automata. Using these models, the spatio-temporal organization of ants on the trail is studied. Based on this, some unusual flow characteristics which differ from those known from other traffic systems, like vehicular traffic or pedestrians dynamics, are found. The theoretical investigations are supplemented by an empirical study of bidirectional traffic on a trail of Leptogenys processionalis. Finally, we discuss some plausible implications of our observations from the perspective of flow optimization.

An Agent-Based Approach to Self-organized Production

Abstract: The chapter describes the modeling of a material handling system with the production of individual units in a scheduled order. The units represent the agents in the model and are transported in the system which is abstracted as a directed graph. Since the hindrances of units on their path to the destination can lead to inefficiencies in the production, the blockages of units are to be reduced. Therefore, the units operate in the system by means of local interactions in the conveying elements and indirect interactions based on a measure of possible hindrances. If most of the units behave cooperatively (“socially”), the blockings in the system are reduced.

Ant clustering with locally weighted ant perception and diversified memory

Abstract: Ant clustering algorithms are a robust and flexible tool for clustering data that have produced some promising results. This paper introduces two improvements that can be incorporated into any ant clustering algorithm: kernel function similarity weights and a similarity memory model replacement scheme. A kernel function weights objects within an ant’s neighborhood according to the object distance and provides an alternate interpretation of the similarity of objects in an ant’s neighborhood. Ants can hill-climb the kernel gradients as they look for a suitable place to drop a carried object. The similarity memory model equips ants with a small memory consisting of a sampling of the current clustering space. We test several kernel functions and memory replacement schemes on the Iris, Wisconsin Breast Cancer, and Lincoln Lab network intrusion datasets. Compared to a basic ant clustering algorithm, we show that kernel functions and the similarity memory model increase clustering speed and cluster quality, especially for datasets with an unbalanced class distribution, such as network intrusion.

Organic Computing and Swarm Intelligence

Abstract: The relations between swarm intelligence and organic computing are discussed in this chapter. The aim of organic computing is to design and study computing systems that consist of many autonomous components and show forms of collective behavior. Such organic computing systems (OC systems) should possess self-x properties (e.g., self-healing, self-managing, self-optimizing), have a decentralized control, and be adaptive to changing requirements of their user. Examples of OC systems are described in this chapter and two case studies are presented that show in detail that OC systems share important properties with social insect colonies and how methods of swarm intelligence can be used to solve problems in organic computing.