Swarm Members Research Articles

Swarming of heterogeneous multi-agent systems with periodically intermittent control

This paper investigates a class of heterogeneous swarming systems with periodically intermittent control. It is assumed that agents in the network are nonidentical and potential functions are heterogeneous. Each agent is assumed to obtain information from the leader and the neighbors only on a series of periodically time intervals. The dynamics of the swarm members are affected by inter-individual interactions and the environment. In addition, we consider the first-order integrator system with a class of attraction/repulsion functions. Some sufficient conditions are provided to guarantee exponential stability of the whole system. Although the information from the leader agent and the neighbors is not continuous, all follower agents can track the leader agent in certain error range. A numerical example is shown to illustrate the validity of the above theoretical result.

Distributed strategy for robots recruitment in swarm-based systems

In a scenario wherein several tasks must be done cooperatively by a robotic swarm, it is sometimes mandatory to prescribe the robots that should execute each one of those tasks. A problem related to the division of labour among the swarm members is the robot recruitment. When a specific robot needs assistance, it must create a group, selecting other robots to help in the execution of the given task. The main contribution of this work is the development of a distributed recruitment strategy based mainly on the propagation of messages among neighbouring robots. The algorithm is evaluated through simulations, wherein some recruitment features are analysed by subjecting the swarm to different conditions, such as the total number of robots in the swarm and the number of robots that start the recruitment. This strategy is promising, since robots have no global knowledge about the swarm, relying only in the neighbourhood information.

Chaotic patterns in the discrete-time dynamics of social foraging swarms with attractant–repellent profiles: an analysis

This article investigates the chaotic dynamical characteristics of a discrete-time social foraging swarm model in which the individuals move in a d-dimensional space according to an attractant/repellent or a nutrient profile. The collective behavior of the swarm model results from the balance between inter-individual interactions and the simultaneous interactions of the swarm members with their environment. Analysis undertaken in the paper indicates that chaos can be introduced into the system by tuning the parameters of the mutual attraction-repulsion function and the attractant–repellent profiles of the swarm dynamics [as given in Eq. (1)]. Ranges of different parameters to ensure sufficient condition for chaotic characteristics have been determined. Apart from chaos, stable and limit cyclic behaviors are also demonstrated for various parameter ranges. Effect of different attractant/repellent gradient profiles on the mean swarm trajectory in the presence of chaos has also been studied. Presence of chaos is determined by means of the Lyapunov exponent. Results obtained from the theoretical analysis are validated through comprehensive simulations.

Large and Dense Swarms: Simulation of a Shortest Path Alarm Propagation

This paper deals with the transmission of alarm messages in large and dense underwater swarms of Autonomous Underwater Vehicles (AUVs) and describes the verification process of the derived algorithm results by means of two simulation tools realized by the authors. A collision-free communication protocol has been developed, tailored to a case where a single AUV needs to send a message to a specific subset of swarm members regarding a perceived danger. The protocol includes a handshaking procedure that creates a silence region before the transmission of the message obtained through specific acoustic tones out of the normal transmission frequencies or through optical signals. This region will include all members of the swarm involved in the alarm message and their neighbours, preventing collisions between them. The AUV sending messages to a target area computes a delay function on appropriate arcs and runs a Dijkstra-like algorithm obtaining a multicast tree. After an explanation of the whole building of this collision-free multicast tree, a simulation has been carried out assuming different scenarios relevant to swarm density, signal power of the modem and the geometrical configuration of the nodes.

Superdiffusive Dispersion and Mixing of Swarms

A common swarm task is to disperse evenly through an environment from an initial tightly packed formation. Due to communication and sensing limitations, it is often necessary to execute this task with little or no communication between swarm members. Unfortunately, prior approaches based on repulsive forces or uniform random walks can often converge quite slowly. With an appropriate choice of random distribution, however, it is possible to generate optimal or near-optimal dispersion and mixing in swarms with zero communication. In particular, we discuss three extremely simple algorithms: reactive Levy walk, reactive ball dispersion, and purely reactive dispersion. All three algorithms vastly outperform prior approaches in both constrained and unconstrained environments, providing a range of options for trading off between aggressiveness and evenness in dispersion.

Time-variant feedback processes in collective decision-making systems: influence and effect of dynamic neighborhood sizes

Self-organizing systems rely on positive feedback (amplification of perturbations). In particular, in swarm systems, positive feedback builds up in a transient phase until maximal positive feedback is reached and the system converges temporarily on a state close to consensus. We investigate two examples of swarm systems showing time-variant positive feedback: alignment in locust swarms and adaptive aggregation of swarms. We identify an influencing bias in the spatial distribution of agents compared to a well-mixed distribution and two features, percentage of aligned swarm members and neighborhood size, that allow us to model the time variance of feedbacks. We report an urn model that is capable of qualitatively representing all these relevant features. The increase in neighborhood sizes over time enables the swarm to lock in a highly aligned state but also allows for infrequent switching between lock-in states. We report similar occurrences of time-variant feedback in a second collective system to indicate the potential for generality of this phenomenon. Our study is concluded by applications of methods from renormalization group theory that allow us to focus on the neighborhood dynamics as scale transformations. Correlation lengths and critical exponents are determined empirically.

Exogenous Fault Detection and Recovery for Swarm Robotics

A robotic swarm needs to maintain continuous operation even in the event of failure of one or more individual robots. Even with a small number of faulty robots, the time for achieving the task will be significantly increased. If the behaviors of biological systems are considered, a number of approaches based on communication, to detect damaged or injured cells or insects can be identified. In this research the transmission of position data between robots has been used as the main method of fault location. The transmission of this data permits members of a robotic swarm to identify and subsequently isolate a faulty robot, either in its drive train or position measurement system. This paper discusses the approach used and presents simulation results.

Self-focusing therapeutic gene delivery with intelligent gene vector swarms: intra-swarm signalling through receptor transgene expression in targeted cells.

Gene delivery in vivo that is tightly focused on the intended target cells is essential to maximize the benefits of gene therapy and to reduce unwanted side-effects. Cell surface markers are immediately available for probing by therapeutic gene vectors and are often used to direct gene transfer with these vectors to specific target cell populations. However, it is not unusual for the choice of available extra-cellular markers to be too scarce to provide a reliable definition of the desired therapeutically relevant set of target cells. Therefore, interrogation of intra-cellular determinants of cell-specificity, such as tissue-specific transcription factors, can be vital in order to provide detailed cell-guiding information to gene vector particles. An important improvement in cell-specific gene delivery can be achieved through auto-buildup in vector homing efficiency using intelligent 'self-focusing' of swarms of vector particles on target cells. Vector self-focusing was previously suggested to rely on the release of diffusible chemo-attractants after a successful target-specific hit by 'scout' vector particles. I hypothesize that intelligent self-focusing behaviour of swarms of cell-targeted therapeutic gene vectors can be accomplished without the employment of difficult-to-use diffusible chemo-attractants, instead relying on the intra-swarm signalling through cells expressing a non-diffusible extra-cellular receptor for the gene vectors. In the proposed model, cell-guiding information is gathered by the 'scout' gene vector particles, which: (1) attach to a variety of cells via a weakly binding (low affinity) receptor; (2) successfully facilitate gene transfer into these cells; (3) query intra-cellular determinants of cell-specificity with their transgene expression control elements and (4) direct the cell-specific biosynthesis of a vector-encoded strongly binding (high affinity) cell-surface receptor. Free members of the vector swarm loaded with therapeutic cargo are then attracted to and internalized into the intended target cells via the expressed cognate strongly binding extra-cellular receptor, causing escalation of gene transfer into these cells and increasing the copy number of the therapeutic gene expression modules. Such self-focusing swarms of gene vectors can be either homogeneous, with 'scout' and 'therapeutic' members of the swarm being structurally identical, or, alternatively, heterogeneous (split), with 'scout' and 'therapeutic' members of the swarm being structurally specialized. It is hoped that the proposed self-focusing cell-targeted gene vector swarms with receptor-mediated intra-swarm signalling could be particularly effective in 'top-up' gene delivery scenarios, achieving high-level and sustained expression of therapeutic transgenes that are prone to shut-down through degradation and silencing. Crucially, in contrast to low-precision 'general location' vector guidance by diffusible chemo-attractants, ear-marking non-diffusible receptors can provide high-accuracy targeting of therapeutic vector particles to the specific cell, which has undergone a 'successful cell-specific hit' by a 'scout' vector particle. Opportunities for cell targeting could be expanded, since in the proposed model of self-focusing it could be possible to probe a broad selection of intra-cellular determinants of cell-specificity and not just to rely exclusively on extra-cellular markers of cell-specificity. By employing such self-focusing gene vectors for the improvement of cell-targeted delivery of therapeutic genes, e.g., in cancer therapy or gene addition therapy of recessive genetic diseases, it could be possible to broaden a leeway for the reduction of the vector load and, consequently, to minimize undesired vector cytotoxicity, immune reactions, and the risk of inadvertent genetic modification of germline cells in genetic treatment in vivo.

Electric circuits performing the swarm optimization

The swarm-based algorithms can be modelled, under suitable assumptions, as equivalent dynamic circuits reproducing the cinematic characteristics of the trajectories followed by the swarm members. This can be made in terms of voltages measured at the terminal of capacitors and currents measured at the terminals of inductors. Through the use of swarm circuits, the role played by the parameters becomes clear since it is possible to apply the stability analysis of continuous systems. This allows us to govern the exploration and/or the exploitation properties of the system simply by tuning its parameters into the convergence range or vice versa. The presented circuital model has been tested on famous benchmarks for optimization and inverse problems. The obtained results show that the swarm circuits are capable to manage exploration as well exploitation and can be used for real-time optimizations such as navigation in unknown ambient of mobile robots and so on.

Levels of Automation for Human Influence of Robot Swarms

Autonomous swarm algorithms and human-robot interaction (HRI) have both attracted increasing attention from researchers in recent years. However, HRI has rarely extended beyond single robots or small multi-robot teams. While one of the benefits of robot swarms is their robust capabilities and the ability of their distributed algorithms to deal autonomously with the complex interactions amongst swarm members, there is undoubtedly a need for humans to influence such swarms in some circumstances—especially when these swarms are operating in unknown or hostile environments. In this paper, we approach the problem of human-swarm interaction (HSI) using previous research in levels of automation (LOAs) in HRI. We create a target searching task whereby the swarm can operate at two different levels of autonomy: an autonomous dispersion algorithm, or user-defined goto points. We investigate what environmental conditions are conducive to different amounts of human influence, and at what point further human intervention has a detrimental effect on the swarm’s performance. The results show that for complex environments containing numerous obstacles and small passageways, there is indeed a need for some human influence; however, after a certain point, further influence causes performance degradation.

MARTIAN SWARM EXPLORATION AND MAPPING USING LASER SLAM

Abstract. In order to explore planet Mars in detail and search for extra-terrestrial life the observation from orbit is not sufficient. To realize complex exploration tasks the use of automatic operating robots with a robust fault-tolerant method of navigation, independent of any infrastructure is a possibility. This work includes a concept of rotary-wing Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) for Martian exploration in a swarm. Besides the scenario of Martian surrounding, with a small number of distinctive landmarks, the challenge consists of a Simultaneous Localization and Mapping (SLAM) concept using laser data of all swarm members.

Stochastic weight trade-off particle swarm optimization for nonconvex economic dispatch

This paper proposes a stochastic weight trade-off particle swarm optimization (SWT_PSO) for solving nonconvex economic dispatch. The proposed SWT_PSO preserves the balance between global exploration and local exploitation to improve the algorithm search capabilities. The balance is retained through trading off stochastic weights amongst previous velocity momentum, cognitive and social components together with using dynamic acceleration coefficients trade-off. The mechanisms for increasing diversity of swarm members are also incorporated to avoid premature convergence. In addition, the novel stochastic trade-off momentum control factor is exploited to enhance the capability of refining quality of a candidate solution during the late search process. The proposed SWT_PSO is tested on four economic dispatch test systems. Test results demonstrate that the proposed approach yields better solution quality than the best reported results in the literature for all test systems.

Adaptive cooperative particle swarm optimizer

An Adaptive Cooperative Particle Swarm Optimizer (ACPSO) is introduced in this paper, which facilitates cooperation technique through the usage of the Learning Automata (LA) algorithm. The cooperative strategy of ACPSO optimizes the problem collaboratively and evaluates it in different contexts. In the ACPSO algorithm, a set of learning automata associated with dimensions of the problem are trying to find the correlated variables of the search space and optimize the problem intelligently. This collective behavior of ACPSO will fulfill the task of adaptive selection of swarm members. Simulations were conducted on four types of benchmark suites which contain three state-of-the-art numerical optimization benchmark functions in addition to one new set of active coordinate rotated test functions. The results demonstrate the learning ability of ACPSO in finding correlated variables of the search space and also describe how efficiently it can optimize the coordinate rotated multimodal problems, composition functions and high-dimensional multimodal problems.

CFSO3: A New Supervised Swarm-Based Optimization Algorithm

We present CFSO3, an optimization heuristic within the class of the swarm intelligence, based on a synergy among three different features of the Continuous Flock-of-Starlings Optimization. One of the main novelties is that this optimizer is no more a classical numerical algorithm since it now can be seen as a continuous dynamic system, which can be treated by using all the mathematical instruments available for managing state equations. In addition, CFSO3allows passing from stochastic approaches to supervised deterministic ones since the random updating of parameters, a typical feature for numerical swam-based optimization algorithms, is now fully substituted by a supervised strategy: in CFSO3the tuning of parameters isa prioridesigned for obtaining both exploration and exploitation. Indeed the exploration, that is, the escaping from a local minimum, as well as the convergence and the refinement to a solution can be designed simply by managing the eigenvalues of the CFSO state equations. Virtually in CFSO3, just the initial values of positions and velocities of the swarm members have to be randomly assigned. Both standard and parallel versions of CFSO3together with validations on classical benchmarks are presented.

Trailing (L5) Neptune Trojans: 2004 KV18 and 2008 LC18

The population of Neptune Trojans is believed to be bigger than that of Jupiter Trojans and that of asteroids in the main belt, although only eight members of this distant asteroid swarm have been observed up to now. Six leading Neptune Trojans around the Lagrange point L4 discovered earlier have been studied in detail, but two trailing ones found recently around the L5 point, 2004 KV18 and 2008 LC18, have not yet been investigated. We report our investigations on the dynamical behaviors of these two new Neptune Trojans. Our calculations show that the asteroid 2004 KV18 is a temporary Neptune Trojan. Most probably, it was captured into the trailing Trojan cloud no earlier than 2.03 × 105 yr ago, and it will not maintain this position later than 1.65 × 105 yr in the future. Based on the statistics from our orbital simulations, we argue that this object is more like a scattered Kuiper belt object. By contrast, the orbit of 2008 LC18 is much more stable. Among the clone orbits spreading within the orbital uncertainties, a considerable portion of clones may survive on the L5 tadpole orbits for 4 Gyr. The strong dependence of the stability on the semimajor axis and resonant angle suggests that further observations are badly required to constrain the orbit in the stable region. We also discuss the implications of the existence and dynamics of these two trailing Trojans over the history of the solar system.

Dynamic Swarm Artificial Bee Colony Algorithm

Artificial Bee Colony (ABC) optimization algorithm is relatively a simple and recent population based probabilistic approach for global optimization. ABC has been outperformed over some Nature Inspired Algorithms (NIAs) when tested over test problems as well as real world optimization problems. This paper presents an attempt to modify ABC to make it less susceptible to stick at local optima and computationally efficient. In the case of local convergence, addition of some external potential solutions may help the swarm to get out of the local valley and if the algorithm is taking too much time to converge then deletion of some swarm members may help to speed up the convergence. Therefore, in this paper a dynamic swarm size strategy in ABC is proposed. The proposed strategy is named as Dynamic Swarm Artificial Bee Colony algorithm (DSABC). To show the performance of DSABC, it is tested over 16 global optimization problems of different complexities and a popular real world optimization problem namely Lennard-Jones potential energy minimization problem. The simulation results show that the proposed strategies outperformed than the basic ABC and three recent variants of ABC, namely, the Gbest-Guided ABC, Best-So-Far ABC and Modified ABC.

An efficient particle swarm optimization technique with chaotic sequence for optimal tuning and placement of PSS in power systems

This paper presents a novel and efficient method for the optimal tuning and placement of power system stabilizer (PSS). The proposed modified particle swarm optimization (MPSO) integrates the particle swarm optimization with passive congregation (PSOPC) and the chaotic sequence. Passive congregation helps each swarm member in receiving a multitude of information from other members and thus decreases the possibility of a failed attempt at detection or a meaningless search. In addition, the chaotic sequence concept is introduced to improve the global searching capability and prevent the premature convergence due to local minima. The proposed optimization procedure handles the problem-specific constraints, in this case it is related to low frequency oscillations in power systems and PSS, using a penalty function. Based on this, tuning and placement of PSS over a wide range of system configurations is formulated as a multi-objective function where the main objective is the aggregation of the three objectives related to the damping ratio and damping factor, and the number of PSS. The robustness of the proposed PSS tuning technique is verified on a multi-machine power system under different operating conditions. The performance of the proposed MPSO is also compared to the PSOPS and PSO, genetic algorithm through eigenvalue analysis, nonlinear time-domain simulation and statistical tests. Finally, the proposed method is applied to find out the best candidate machines to be equipped with PSSs. The obtained results showed that the new method can find the best placements and the optimum PSSs parameters simultaneously with an excellent global damping performance.

Direction election in flocking swarms

Swarm formation and swarm flocking may conflict each other. Without explicit communication, such conflicts may lead to undesired topological changes since there is no global signal that facilitates coordinated and safe switching from one behavior to the other. Moreover, without coordination signals multiple swarm members might simultaneously assume leadership, and their conflicting leading directions are likely to prevent successful flocking. To the best of our knowledge, we present the first set of swarm flocking algorithms that maintain connectivity while electing direction for flocking, under conditions of no communication. The algorithms allow spontaneous direction requests and support direction changes.

Mineralogical and Geochemical Controls on the Formation of the Woods Point Dike Swarm, Victoria, Australia: Evidence from the Morning Star Dike and Implications for Sourcing of Au Within Orogenic Gold Systems

The Woods Point dike swarm of eastern Victoria, Australia, has produced over 133 tonnes (t) of Au. Here, as in a number of other orogenic Au provinces, a close spatial relationship is observed between orogenic Au mineralization and intrusive rocks; this study focuses on the well-endowed Morning Star dike, a gabbro to gabbro-diorite dike that has produced ~28 t of hydrothermal quartz-carbonate-pyrite vein-hosted Au and a further 24 t of placer gold derived from primary rocks in the area immediately around Morning Star. The fractionated magma that formed the dikes within the Woods Point dike swarm was sourced from a staging magma chamber at a depth of 16 to 23 km; here, mantle-derived magmas were driven to S saturation by assimilation of crustal material, leading to the formation of Cu-Ni-platinum group element (PGE)-Au–bearing sulfides, most of which segregated and settled toward the bottom of the staging chamber. Emplacement of the dike initiated with tapping of the staging chamber and release of the magmas within the chamber, forming the Woods Point dike swarm; flow differentiation and the formation of dike bulges enabled the segregation of sulfide-bearing mafic units and intermediate sulfide-barren units within these dikes. This is exemplified by primitive amphibole-rich, sulfide-bearing gabbroic sections and more evolved sulfide-barren gabbro-diorite portions of the Morning Star dike. We suggest that the source of Au within the Morning Star dike as well as that in other members of the Woods Point dike swarm was the Cu-PGE-Au sulfides left at depth within the magma conduit system, for example within the Thomson River Copper Mine dike. Mass balance calculations undertaken suggest that ~104 million tonnes (Mt) of magmatic sulfide may exist within the magma conduit system; these magmatic sulfides, if they are the equivalent of the sulfides preserved in the Morning Star dike, probably contain around 3% Ni and 10% Cu. During postmagmatic metamorphism, we infer that hydrothermal fluids flowed along dikes, interacting with magmatic sulfides and leaching some of the Au, which was subsequently deposited in quartz-carbonate veins at higher levels in the system.

Power system stabilizer design using hybrid multi-objective particle swarm optimization with chaos

A novel technique for the optimal tuning of power system stabilizer (PSS) was proposed, by integrating the modified particle swarm optimization (MPSO) with the chaos (MPSOC). Firstly, a modification in the particle swarm optimization (PSO) was made by introducing passive congregation (PC). It helps each swarm member in receiving a multitude of information from other members and thus decreases the possibility of a failed attempt at detection or a meaningless search. Secondly, the MPSO and chaos were hybridized (MPSOC) to improve the global searching capability and prevent the premature convergence due to local minima. The robustness of the proposed PSS tuning technique was verified on a multi-machine power system under different operating conditions. The performance of the proposed MPSOC was compared to the MPSO, PSO and GA through eigenvalue analysis, nonlinear time-domain simulation and statistical tests. Eigenvalue analysis shows acceptable damping of the low-frequency modes and time domain simulations also show that the oscillations of synchronous machines can be rapidly damped for power systems with the proposed PSSs. The results show that the presented algorithm has a faster convergence rate with higher degree of accuracy than the GA, PSO and MPSO.