Guest editorial

Citation

Hettiarachchi, S. and Spears, W.M. (2009), "Guest editorial", International Journal of Intelligent Computing and Cybernetics, Vol. 2 No. 4. https://doi.org/10.1108/ijicc.2009.39802daa.001

Guest editorial

Article Type: Guest editorial From: International Journal of Intelligent Computing and Cybernetics, Volume 2, Issue 4

About the Guest Editors

Suranga Hettiarachchi received his PhD in Computer Science from the University of Wyoming in 2007. He is currently an Assistant Professor at Indiana University Southeast and is also the Founder of the Swarm Robotics Laboratory there. His research interests include milti-agent systems, physicomimetics, distributed robotics, evolutionary learning, search algorithms, adaptive learning, and swarm intelligence. His publications have appeared in refereed international journals and conferences and this is his first co-editorial effort.

William M. Spears received a PhD in Computer Science from George Mason University in 1998. He has an international reputation for his expertise in evolutionary computing and has a published book on the topic. He has co-edited books on swarm robotics and evolutionary computation. His current research includes distributed robotics, the epidemiology of virus spread, evolutionary algorithms, complex adaptive systems, and learning and adaptation. He was co-founder of the University of Wyoming Distributed Robotics Laboratory and has approximately 75 publications. He is the CEO of Swarmotics, LLC.

The focus of swarm robotics is to study how a swarm of relatively simple physically embodied robots can be designed and controlled to collectively accomplish tasks that are beyond the capabilities of a single robot. The advantages of swarm robotics are that: complex control is achievable through simple local interactions of the swarm members; the results scale well with larger numbers of robots; and the swarm is robust to withstand failure of individual members. Algorithms, techniques, and methods based on swarm robotics principles have been successfully applied to a wide range of complex problems. This special issue aims at exhibiting the latest research achievements, findings and ideas, theoretical as well as practical, in the area of swarm robotics.

The nine papers presented in this special issue provide original research findings with in-depth analysis. Given the relative novelty of swarm robotics, our intention was to provide the reader with a diverse collection of research from many different areas of swarm robotics, rather than focus on a particular approach or problem area. The papers presented in this special issue combine multiple techniques that use multiple approaches. Contributions include:

• •

  the use of a model for swarm foraging with temporal logic that is constrained by real time behavior;

• •

  the combination of physicomimetics with distributed evolutionary learning;

• •

  the development of a novel swarm engineering model;

• •

  the presentation of a new recombination operator in evolutionary robotics that narrows down the dimensions of the search space;

• •

  the combination of a neuro-fuzzy system with a reinforcement learning (RL) algorithm for adaptive swarm behaviors;

• •

  a new approach for chemical plume tracing (CPT) based on the physics of fluid dynamics;

• •

  a mathematical model for swarm pattern formations and transformations;

• •

  the examination of performance of different map merging strategies for simultaneous localization and mapping problems; and

• •

  the presentation of a technique for the coordination of swarm robots with low capabilities driven by instructions learned from on-site radio frequency identification (RFID) tags.

As can be seen, given the newness of the field, it is difficult to group these papers into traditional categories. Indeed, we prefer to allow the field to mature and to allow the categories to self-emerge. In fact, the papers may even differ in their precise definition of what constitutes a swarm. We do not view this as a weakness, but as an acknowledgement that the field is still in an exploratory stage. In addition to diversity in the techniques, the papers in this special issue also present practical implications and theoretical boundaries of several swarm robotics techniques. We expect that the robust and multi-disciplinary research techniques presented in this special issue would definitely allow the reader to enhance the quality of their own research. The papers presented in this special issue of swarm robotics are as follows.

The first paper entitled “Deductive verification of simple foraging robotic behaviors” authored by “Abdelkader Behdenna, Clare Dixon, and Michael Fisher” presents a logical specification, and automated verification of high-level robotic behaviors using temporal logic as a formal language for providing abstractions. They achieve many properties using automatic deductive temporal provers for the propositional and first-order temporal logics.

The second paper entitled “Distributed adaptive swarm for obstacle avoidance” authored by “Suranga Hettiarachchi and William M. Spears” demonstrates a novel use of a new force law in physicomimetics that allows a swarm of robots to surpass the prior state-of-the-art in obstacle avoidance tasks. They also present a decentralized online learning approach called “DAEDALUS” for the adaptation of swarms when the environment changes.

The third paper entitled “Model independence in swarm robotics” authored by “S. Kazadi” examines the development of a methodology for generating swarms and illustrates the methodology using lossless flocking. The swarm engineering method is also used to develop a new physicomimetics method, referred to as the “quark” model.

The fourth paper entitled “Decentralized evolution of robotic behavior using finite state machines” authored by “Lukas König, Sanaz Mostaghim, and Hartmut Schmeck” utilizes a control system representation based on finite state machines to build a decentralized online-evolutionary framework for swarms of mobile robots. They narrow down the dimensions of the search space by using a new recombination operator for multi-parental generation of offspring and by extending a known mutation operator that harden parts of genotypes involved in good behavior. The results show that the framework is capable of robustly evolving the benchmark behaviors.

The fifth paper entitled “Adaptive swarm behavior acquisition by a neuro-fuzzy system and reinforcement learning algorithm” authored by “Takashi Kuremoto, Masanao Obayashi, and Kunikazu Kobayashi” presents a neuro-fuzzy system with a RL algorithm for adaptive swarm behaviors. The simulation experiments applies the proposed system on the goal-directed navigation problem and the results show that swarms were successfully formed and optimized routes were found faster than the case of individual learning.

The sixth paper entitled “Foundations of swarm robotic chemical plume tracing from a fluid dynamics perspective” authored by “Diana F. Spears, David R. Thayer, and Dimitri V. Zarzhitsky” provides a foundation for CPT based on the physics of fluid dynamics. They present a theoretical approach for source localization using the divergence theorem of vector calculus, and the fundamental underlying notion of the divergence of the chemical mass flux. Their CPT algorithm, called “fluxotaxis,” follows the gradient of this mass flux to locate a chemical source emitter and is shown to be correct in the sense that it is guaranteed to point toward a true source emitter and not be fooled by fluid sinks.

The seventh paper entitled “A review and implementation of swarm pattern formation and transformation models” authored by “Blesson Varghese and Gerard McKee” addresses the classic problem of pattern formation identified by researchers in the area of swarm robotic systems. They propose a pattern formation model based on mathematical foundations and macroscopic primitives. Their paper introduces a formal definition for swarm pattern transformation and four special cases of transformations. They confirm the validity of the proposed models, and the feasibility of the methods investigated on the Traer Physics and Processing environment.

The eighth paper entitled “Graph exploration with robot swarms” authored by “Hui Wang, Michael Jenkin, and Patrick Dymond” explores the simultaneous localization and mapping problem when large groups of robots operate within the environment. They examine the performance of two different map merging strategies, present a correctness proof of the algorithm, and compare different coordination strategies via simulation. Their experimental results demonstrate that a swarm of identical robots, each equipped with its own marker, and capable of simple sensing and action abilities, can explore and map an unknown graph-like environment.

The last paper entitled “A swarm of robots using RFID tags for synchronization and cooperation” authored by “Giulio Zecca, Paul Couderc, Michel Banâtre, and Roberto Beraldi” focuses on a technique for the coordination of a swarm of robots with low capabilities driven by instructions learned from on-site RFID tags that are used as a pervasive memory distributed in the environment. Their robots exploit ubiquitous computing to make a formation in space, synchronize with team mates in the same zone, and finally complete a cooperative task. They provide a validation of their algorithm through a simulation environment, showing its applicability and performance, before the real implementation on Roomba-like robots.

The call for papers for this special issue received 14 submissions, and each submission was peer-reviewed by at least two experts in the related field. After the revisions are made according to the feedback, nine papers were accepted to appear in this special issue, which includes six papers from the special issue submissions, two papers from the regular submissions, and one invited paper. The guest editors would like to thank the authors and the reviewers for their contributions to this special issue. Moreover, we are grateful to the International Journal of Intelligent Computing and Cybernetics for the opportunity to publish and the journal editors for their insightful feedback to this special issue in swarm robotics, and their support and guidance in the publication.

Suranga Hettiarachchi, William M. SpearsGuest Editors