Belief-Desire-Intention Agent Model Research Articles

On the impact of fuzzy-logic based BDI agent model for cyber–physical systems

Cyber–Physical Systems (CPS) interconnect embedded computing technologies into the physical world, forming a complex, multi-disciplinary, physically-effective system. However, interacting with the physical world brings unpredictability, as real-world events are uncertain and dynamic by its nature. Consequently, CPS must be capable of reasoning about these unpredictable situations and adapt their behaviour accordingly. Therefore, it is essential to enhance the intelligence of CPS to overcome this challenge effectively, leading to the development of smart CPS. As an approach, a reasoning mechanism can support the system to reason about its state and actions to handle unpredictable changes. In this study, we employ intelligent Belief-Desire-Intention (BDI) agents to achieve this objective. Nevertheless, traditional logic approaches based on crisp numbers, which are used in typical BDI agents, may not effectively handle uncertainty. Hence, the reasoning mechanism can be extended with fuzzy logic to deal with run-time uncertainties. Thus, the target system’s suitable inputs, outputs and reasoning phases are fuzzified to indicate the impact of the fuzzy logic on the CPS. To this end, an architecture is provided to deploy BDI agents integrated with a fuzzy reasoning model for handling imprecise information and improving process control. The approach is validated and evaluated on a complex case study with a heterogeneous structure controlled by multiple agents, namely the “smart production system”. The multiple modular and end-to-end experiments conducted reveal that, on the whole, the fuzzy BDI agent outperforms the classical one by up to three times, with only requiring approximately 10% more computation time.

CBDI: Towards an Architecture for Human–Machine Collaboration

Effective human–machine collaboration is a long standing problem within rehabilitation robotics. Machines that affect peer-to-peer interaction hold promise. Agents can serve as the basis for such machines. Belief–desire–intention (BDI) paradigm is used to achieve human-like intelligence. However, the BDI model does not emulate human-like qualities such as coordination or collaborative planning. Therein lies the motivation to extend the basic BDI agent model. cBDI—an extension to the BDI architecture supporting human–machine collaboration is presented. Two different applications for the cBDI agent are highlighted: (1) human–cBDI agent collaboration to achieve block stacking and (2) cBDI-based collaborative navigation for an intelligent wheelchair (IW). Under scenarios such that the block stacking task cannot be completed by either human or agent alone, we show how the cBDI is tuned to collaborate and accomplish the task. The cBDI-based collaborative navigation controller is for affecting assistance-as-required in an IW. The effectiveness of the cBDI-based collaborative architecture is evaluated within a ROS-USARSim framework.

DESIGNING A BDI AGENT REACTANT MODEL OF BEHAVIOURAL CHANGE INTERVENTION

Belief-Desire-Intention (BDI) model is well suited for describing agent’s mental state. The BDI of an agent represents its motivational stance and are the main determinant of agent’s actions. Therefore, explicit understanding of the representation and modelling of such motivational stance plays a central role in designing BDI agent with successful behavioural change interventions. Nevertheless, existing BDI agent models do not represent agent’s behavioural factors explicitly. This leads to a gap between design and implementation where psychological reactance has being identified as the cause of BDI agent behavioural change interventions failure. Hence, this paper presents a generic representation of BDI agent model based on behavioural change and psychological theories. Also, using mathematical analysis the model was evaluated. The objective of the proposed BDI agent model is to bridge the gap between agent design and implementation for successful agent-based interventions. The model will be realized in an agent-based application that motivates children towards oral hygiene. The study explicitly depicts how agent’s behavioural factors interact to enhance behaviour change which will assist agent-based intervention designers to be able to design intervention that will be void of reactance.

Translation of statechart agents into a BDI framework for MAS engineering

In the agent-oriented computing area, different programming paradigms have been to date proposed to design and implement multi-agent systems (MAS). Among them, the Belief-Desire-Intention (BDI) and Statecharts-based models have emerged as they effectively support the definition of rational and protocol-oriented agent behaviors, respectively. In this paper, we investigate the translation between such models by using two available agent modeling languages: PROFETA, which is based on the BDI agent model, and the agent-oriented Distilled StateCharts (DSC) formalism, which allows modeling agent behaviors based on Statecharts-like state machines. In particular, we propose a mapping from DSC to PROFETA that can enable visual programming of PROFETA programs to facilitate the development of PROFETA-based autonomous systems. A case-study is also developed to exemplify the proposed engineering approach.

Deployment of an Agent-based SANET Architecture for Healthcare Services

Deployment of an Agent-based SANET Architecture for Healthcare ServicesThis paper describes the adaptation of a computational technique utilizing Extended Kohonen Maps (EKMs) and Rao-Blackwell-Kolmogorov (R-B) Filtering mechanisms for the administration of Sensor-Actuator networks (SANETs). Inspired by the BDI (Belief-Desire-Intention) Agent model from Rao and Georgeff, EKMs perform the quantitative analysis of an algorithmic artificial neural network process by using an indirect-mapping EKM to self-organize, while the Rao-Blackwell filtering mechanism reduces the external noise and interference in the problem set introduced through the self-organization process. Initial results demonstrate that a combinatorial approach to optimization with EKMs and Rao-Blackwell filtering provides an improvement in event trajectory approximation in comparison to standalone cooperative EKM processes to allow responsive event detection and optimization in patient healthcare.

An Object-Oriented Design Model of Software Agent

How to implement software Agent is a key problem to develop Agent-oriented programming languages and development tools. Aiming to solve this problem based on the mainstream OO (object oriented) technology, this paper first discusses the differences between Agent and object, and then presents an implemental architecture and behaviors decision algorithm of software Agent. The architecture and related algorithm are based on OO technology and the simplified improvement of BDI (belief desire intention) Agent model. An OO-based design framework of software Agent is also proposed by using the POAD (pattern-oriented analysis and design) method. The approach is helpful to clarify how to extend the OO method suitably for developing Agent-oriented programming languages and development tools.

A New Way of Discovery of Belief, Desire and Intention in the BDI Agent-Based Software Modeling

Agent-based programming has been emerged as a next generation programming paradigm. There are many different definitions and usage for agents. In our research, however, an agent is defined as an autonomous, concurrent and intelligent object. Furthermore, our agents are modeled by the belief-desire-intention (BDI) concept. An agent is embodied when it is assigned to its BDI. We call it a BDI agent. A software process defines a set of activities and associated artifacts that lead to the construction of a software system. We have developed a software process based on the BDI agent model that is useful for a systematic development of BDI agent-based software construction. We named our process as the BDI Agent-based Software Process (BDI ASP). This paper presents a new way of modeling technique in our BDI ASP. This work will convince us that the BDI ASP is very sound and practicable in agent software construction. We will provide a few examples as a case study with brief explanations of activities and artifacts in our process.