Now showing 1 - 10 of 35
  • Publication
    Dublin Bogtrotters : agent herders
    This paper describes an entry to the Multi-Agent Programming Contest 2008. The approach employs the pre-existing Agent Factory framework and extends this framework in line with experience gained from its use within the robotics domain.
      461Scopus© Citations 3
  • Publication
    EDLA tradeoffs for wireless sensor network target tracking
    The number of active nodes in a WSN deployment governs both the longevity of the network and the accuracy of applications using the network’s data. As node hibernation techniques become more sophistocated, it is important that an accurate evaluation methodology is employed to ensure fair comparisons across different techniques. Examining both energy and accuracy ensures a claim of increased longevity for a particular technique can be contrasted against its associated drop, if any, in application accuracy. This change can also be as a result of increased latency and the accuracy encapsulates many aspects of WSN performance in one metric. In this work, we detail the first in a series of simulation experiments designed to demonstrate the tradeoffs for a WSN and we employ mobility tracking as the application to benchmark accuracy. Additionally, we demonstrate experimental evidence for a potential adaptive mobility tracking protocol.
      2200Scopus© Citations 8
  • Publication
    Towards evolutionary ambient assisted living systems
    Ambient Assisted Living (AAL) is advocated as technological solutions that will enable the elderly population maintain their independence for a longer time than would otherwise be the case. Though the facts motivating the need for AAL are indisputable, the inherently heterogeneous nature and requirements of the elderly population raise significant difficulties. One particular challenge is that of designing AAL systems that can evolve to meet the requirements of individuals as their needs and circumstances change. This demands the availability of an adaptive, open, scalable software platform that incorporates a select combination of autonomic and intelligent techniques. Given that the first generation of AAL systems will be deployed in the near future, it is incumbent on designers to factor this need for evolution and adaptivity in their designs and implementations. Thus this paper explores AAL from a number of prospective and considers an agent-based middleware approach to realising an architecture for evolutionary AAL.
      5999Scopus© Citations 71
  • Publication
    Agent-based coordination for the sensor web
    The approach described advocates the use of a multi-agent system, and specifically the use of multi-agent distributed constraint optimisation algorithms. Developing software for low powered sensing devices introduces several problems to be addressed; the most obvious being the limited computational resources available. In this paper we discuss an implementation of ADOPT, a pre-existing algorithm for distributed constraint optimisation, and describe how it has been integrated with a reflective agent platform developed for resource constrained devices, namely Agent Factory Micro Edition (AFME). The usefulness of this work is illustrated through the canonical multi-agent coordination problem, namely graph colouring.
      16043Scopus© Citations 2
  • Publication
    MERLIN: A synergetic integration of MAC and Routing Protocol for Distributed Sensor Networks
    Notoriously, energy-efficient MAC protocols cause high latency of packets. Such delays may well increase when a routing protocol is applied. Therefore, quantifying the endto- end delay and energy consumption when low duty cycle MAC and routing protocols are jointly used, is of particular interest. In this paper, we present a comprehensive evaluation of the MERLIN (MAC and Efficient Routing integrated with support for localization) protocol. MERLIN integrates MAC and routing features into a single architecture. In contrast to many sensor network protocols, it employs a multicast upstream and multicast downstream approach to relaying packets to and from the gateway. Simultaneous reception and transmission errors are notified by using asynchronous burst ACK and negative burst ACK. A division of the network into timezones, together with an appropriate scheduling policy, enables the routing of packets to the closest gateway. An evaluation of MERLIN has been conducted through simulation, against both the SMAC and the ESR routing protocols,which is an improved version of the DSR algorithm. The results illustrate how both SMAC and ESR, jointly used in low duty cycle scenarios, can cause an impractical and very high end-to-end delays. MERLIN, as an integrated approach, notably reduces the latency, resulting in nodes that can operate in a very low duty cycle. Consequently, an extension of the operative lifetime of the sensor network is achieved.
      500Scopus© Citations 12
  • Publication
    Virtual sensor networks : an embedded agent approach
    Many documented instances of existing research on Wireless Sensor Networks (WSN) use deployments that either fall short of, or barely meet, the resource requirements of the application. In this paper, it is envisaged that future WSN deployments will far exceed the resource requirements of any one single application. In a similar fashion to the use of virtual machines on a mainframe, sub-networks of adequate resources will be carved out of the entire deployment to fulfil the requirements of multiple applications. These will be hosted simultaneously on the network, and in many cases, certain WSN nodes will form a component in a number of these Virtual Sensor Networks (VSN). Such VSNs will also be dynamic in nature, adapting resources as nodes go offline. An additional requirement of such networks will be to engage in opportunistic power management, such as node hibernation, while the networks are adapting. In this paper, a solution for both of these issues is proposed, underpinned by a Multi-Agent System (MAS) resident on individual nodes. This solution facilitates both the practical operation of adaptive VSNs, while ensuring aggregate energy consumption can be minimised.
      2583Scopus© Citations 14
  • Publication
    Embedding intelligent decision making within complex dynamic environments
    Decision-making is a complex and demanding process often constrained in a number of possibly conflicting dimensions including quality, responsiveness and cost. This paper considers in situ decision making whereby decisions are effected based upon inferences made from both locally sensed data and data aggregated from a sensor network. Such sensing devices that comprise a sensor network are often computationally challenged and present an additional constraint upon the reasoning process. This paper describes a hybrid reasoning approach to deliver in situ decision making which combines stream based computing with multi-agent system techniques. This approach is illustrated and exercised through an environmental demonstrator project entitled SmartBay which seeks to deliver in situ real time environmental monitoring.
      1374Scopus© Citations 12
  • Publication
    Implicitly and intelligently influencing the interactive experience
    Enabling intuitive interaction in system design remains an art more than a science. This difficulty is exacerbated when the diversity of device and end user group is considered. In this paper, it is argued that conventional interaction modalities are unsuitable in many circumstances and that alternative modalities need be considered. Specifically the case of implicit interaction is considered, and the paper discusses how its use may lead to more satisfactory experiences. Specifically, harnessing implicit interaction in conjunction with the traditional explicit interaction modality, can enable a more intuitive and natural interactive experience. However, the exercise of capturing and interpreting implicit interaction is problematic and is one that lends itself to the adoption of AI techniques. In this position paper, the potential of lightweight intelligent agents is proposed as a model for harmonising the explicit and implicit components of an arbitrary interaction.
      569
  • Publication
    Autonomous management and control of sensor network-based applications
    A central challenge facing sensor network research and development is the difficulty in providing effective autonomous management capability. This is due to a large number of parameters to control, unexpected changes of the network topology and dynamic application requirements. Network management is also a challenging task for the remote user due to the large-scale of the network and scarce visibility of live network happenings. Preferably the network should have autonomous decision-making capabilities as network conditions and application requirements changes. To cope with such uncertainties, firstly we consider Octopus, a powerful software tool that provides live information about the network topology and sensor data. At present, the tool can provide monitoring and require a user to control the network state manually. This paper describes how Octopus is reengineered to accommodate a multi-agent system to provide autonomic managing capabilities. In particular, we detail two distinct architectures, the static and mobile agent architectures, which can be effectively applied to deliver autonomous system management. This paper sets the basis for a full autonomous network management via a multi agent system to work with Octopus.
      1162Scopus© Citations 2
  • Publication
    Benchmarking latency effects on mobility tracking in WSN
    The number of active nodes in a WSN deployment governs both the longevity of the network and the accuracy of applications using the network’s data. As node hibernation techniques become more sophistocated, it is important that an accurate evaluation methodology is employed to ensure fair comparisons across different techniques. Examining both energy and accuracy ensures a claim of increased longevity for a particular technique can be contrasted against its associated drop, if any, in application accuracy. This change can also be as a result of increased latency and the accuracy encapsulates many aspects of WSN performance in one metric. In this work, we detail the first in a series of experiments designed to demonstrate the tradeoffs for a WSN and we employ mobility tracking as the application to benchmark accuracy. Additionally, we demonstrate experimental evidence for a potential adaptive mobility tracking protocol.
      557Scopus© Citations 3