Belief-Desire-Intention Model Research Articles

Rational software agents with the BDI reasoning model for Cyber–Physical Systems

Cyber–Physical Systems (CPSs) are complex and heterogeneous systems. Interacting with the physical world makes CPS unpredictable because of unanticipated events. Therefore, a CPS needs to reason these events autonomously. Henceforward, suitable reasoning mechanisms, such as rational agents with deliberative capabilities, should be selected and integrated into the CPSs. In this way, the integrated multi-agents can reason on environmental changes to find a plan that sustains the system’s operation for CPS. In addition, a layered architecture can pave the way to integrate the rational agents on embedded hardware and control the CPS. To this end, this paper presents an architecture, discusses a reference implementation and elaborates on the high-level integration of agents and CPS. Moreover, a complex and heterogeneous case study is provided to validate the effectiveness of rational agents in conducted experiments. Firstly, the rational agents utilising the belief–desire–intention (BDI) model required approximately three times less development time than simple-reflex agents. Secondly, the proposed approach resulted in up to 3 times less description complexity in language expressiveness. Lastly, the product quality is improved up to 66% by the rational agents and BDI model. As a result, our approach is beneficial to designing multi-agent CPS where it is aimed to use low-level control and high-level reasoning in a single platform.

Boosting stakeholders' intention to participate in socially responsible collective action in megaprojects: perspectives of mega water transfer projects in China

PurposeMegaprojects have large impact on the environment and stakeholders should take collective action to ensure that these projects are developed in a socially responsible manner. Hitherto, it is not known whether group and subjective norms and social identity could compel stakeholders to take socially responsible collective actions in megaprojects. The aim of this study is to design and test a model to boost stakeholders' intention to take socially responsible collective action in the context of mega water transfer projects in China.Design/methodology/approachA quasi-experimental causal research design was adopted to establish cause–effect relationships among the dependent variable (we-intention) and independent variables (subjective norms, group norms, social identity and desire). This study adopts the belief–desire–intention model and social influence theory to empirically investigate how to boost the stakeholders' intention to participate in socially responsible collective action. An online questionnaire survey was conducted and data was collected from 365 respondents who were involved in mega water transfer projects in China. The partial least squares structural equation modeling technique was employed to analyze the data.FindingsThe results from partial least squares analyses indicate that the presence of subjective norms, group norms and social identity (collectively known as social influence process) could increase stakeholders' intention to take socially responsible collective action. In addition, the desire to be socially responsible also boosts stakeholders' intention to take collective action. Desire partially mediates the relationship between social influence process and intention to take socially responsible collective action.Originality/valueThis study adds to existing knowledge by discovering social influence process as an antecedent to taking socially responsible collective action in megaprojects. Strong group norms and subjective norms could propel stakeholders to be more socially responsible. The study also adds to knowledge by discovering that stakeholders' desire to fulfill social responsibility also leads them to take concrete actions. Implications and recommendations are provided on how to manipulate different types of social influence processes to facilitate stakeholders to adopt socially responsible collective action in the process of managing megaprojects.

Flexible Agent Architecture: Mixing Reactive and Deliberative Behaviors in SPADE

Over the years, multi-agent systems (MAS) technologies have shown their usefulness in creating distributed applications focused on autonomous intelligent processes. For this purpose, many frameworks for supporting multi-agent systems have been developed, normally oriented towards a particular type of agent architecture (e.g., reactive or deliberative agents). It is common, for example, for a multi-agent platform supporting the BDI (Belief, Desire, Intention) model to provide this agent model exclusively. In most of the existing agent platforms, it is possible to develop either behavior-based agents or deliberative agents based on the BDI cycle, but not both. In this sense, there is a clear lack of flexibility when agents need to perform part of their decision-making process according to the BDI paradigm and, in parallel, require some other behaviors that do not need such a deliberation process. In this context, this paper proposes the introduction of an agent architecture called Flexible Agent Architecture (FAA) that supports the development of multi-agent systems, where each agent can define its actions in terms of different computational models (BDI, procedural, neural networks, etc.) as behaviors, and combine these behaviors as necessary in order to achieve its goals. The FAA architecture has been integrated into a real agent platform, SPADE, thus extending its original capabilities in order to develop applications featuring reactive, deliberative, and hybrid agents. The integration has also adapted the existing facilities of SPADE to all types of behaviors inside agents, for example, the coordination of agents by using a presence notification mechanism, which is a unique feature of SPADE. The resulting SPADE middleware has been used to implement a case study in a simulated robotics scenario, also shown in the paper.

Features of Decision-Making in the Process of lending to small businesses: the bDi Model and trust indices

The aim of the study is to identify the problems hindering the effective interaction of banks with SMEs and to determine the role of trust and other institutional factors in decision-making. Despite the recovery in the SME lending sector in 2019–2020, the needs of small firms in loans are not fully satisfied. The task of enhancing interactions between banks and the sectors under study requires a fundamental approach and is relevant.The BDI (Belief-DesireIntention) model developed by M. Bratman was used as a method for investigating the decision-making process in the subject area under consideration. The place of this tool in the system of methods of analysis and support of decisionmaking processes is shown; the universality of the BDI model is substantiated and the main areas of its application are indicated.The scientific novelty of the paper is the modification of the BDI model, which is carried out by taking into account the trust factor and increasing the number of decision-making phases.The prospects of using the modified BDI model as a methodological basis for the formation of the architecture of banking ecosystems are considered. On the example of the experience of a quantitative assessment of the trust factor between banks and small firms, the possibilities of using the index method to measure other institutional factors of the modified M. Bratman model and determine the level of feasibility of a transaction planned in the area of lending are investigated.The author concluded that a necessary condition for enhancing the lending activities of banks interacting with small firms is the formation of an adequate institutional environment in accordance with the methodological principles outlined in this article.

Semantic Interaction Strategy of Multiagent System in Large-Scale Intelligent Sensor Network Environment

In a multiagent system, the semantic interaction between agents is an important aspect affecting multi-intelligence. The purpose of interaction is to reasonably arrange task objectives and behaviors through information sharing and communication interaction, so as to maximize the overall performance of multiagent system. This paper analyzes the communication and interaction process between agents from the perspective of semantic layer and introduces the BDI (belief, desire, intention) model of agent’s thinking state into the communication and interaction process. Furthermore, we propose a multiagent semantic interaction strategy model based on a large-scale intelligent sensor network, which supports various types of negotiation and interaction on the basis of basic interaction behavior to solve the problem of information operational conflicts. In addition, this paper limits the scale of historical information through the definition of equivalence and the merging theorem of history, and it uses reinforcement learning algorithm to detect possible conflicts and delay communication and makes rational use of limited resources to improve system revenue and coordination efficiency. The experimental results show that compared with the previous methods such as debate and negotiation, the strategy model can realize the flexible interaction based on scene and is more practical. At the same time, the existence of reinforcement learning improves the efficiency analysis and the convergence performance of semantic interaction strategy.

Behavioral flexibility in Belief-Desire- Intention (BDI) architectures

The Belief-Desire-Intention (BDI) model is a popular approach to design flexible agents. The key ingredient of BDI model, that contributed to concretize behavioral flexibility, is the inclusion of the practical reasoning. On the other hand, researchers signaled some missing flexibility’s ingredient, in BDI model, essentially the lack of learning. Therefore, an extensive research was conducted in order to extend BDI agents with learning. Although this latter body of research is important, the key contribution of BDI model, i.e., practical reasoning, did not receive a sufficient attention. For instance, for performance reasons, some of the concepts included in the BDI model are neglected by BDI architectures. Neglecting these concepts was criticized by some researchers, as the ability of the agent to reason will be limited, which eventually leads to a more or less flexible reasoning, depending on the concepts explicitly included. The current paper aims to stimulate the researchers to re-explore the concretization of practical reasoning in BDI architectures. Concretely, this paper aims to stimulate a critical review of BDI architectures regarding the flexibility, inherent from the practical reasoning, in the context of single agents, situated in an environment which is not associated with uncertainty. Based on this review, we sketch a new orientation and some suggested improvements for the design of BDI agents. Finally, a simple experiment on a specific case study is carried out to evaluate some suggested improvements, namely the contribution of the agent’s “well-informedness” in the enhancement of the behavioral flexibility.

Proposition of a Distributed Voronoi Partitioning Approach Enhanced with a Dispersion Phase for a Multirobot System

This paper is an extension to our work presented in Ben Slimane and Tagina (in: Nguyen, Pimenidis, Khan and Trawinski (eds) Computational collective intelligence, Springer, Cham, 2018). It deals with the the problem of partitioning the space in an even way between a number of autonomous mobile robots. In our previous work we proposed a distributed one-phased partitioning method where each robot constructs its corresponding Voronoi cell from the information received from its neighbors. We propose in what follows a two-phased partitioning approach, starting with a dispersion task, followed by the distributed Voronoi partitioning as for the one-phased method. For the dispersion phase, we propose a novel parametrized algorithm from which we seek to control the dispersion behavior of the robots. The individual actions of the agents are controlled by the belief–desire–intention model which endows them with the required know-how needed to operate deliberately and readjust the plans dynamically on the go. We show in this paper, through a series of experiments, the results of the dispersion method and the impact of its parameters on the generated maps. We also compare the results of the two partitioning methods to show the impact of the dispersion on the partitioning in terms of the actual performed steps towards convergence and the generated maps for both methods.

"Editorial Reflections: Lockdowns, Slow Downs, and Some Introductions "

To qualify as a JAIS AE, an individual must be an associate-level (or the equivalent) faculty member and must have published at least two publications in such Journals as JAIS, MIS Quarterly, Information Systems Research, and the Journal of Management Information Systems The second research article-Reinforcing Effects of Formal Control Enactment in Complex IT Projects by Gloria Liu and Cecil Chua-is a case study of a large manufacturing organization that designs and manufactures wireless communication products embedding software and hardware to enable new product functionality In the third research article in this issue-Predicting Intention to Participate in Socially Responsible Collection Action in a Social Networking Website Group by Victor Chen, Timothy Hiele, Adam Kryszak, and William Ross-the authors discuss the notion of we-intention, which highlights individuals' commitment to group activities Drawing on several theoretical concepts from the belief-desire-intention model and social influence theory and data from a sample of Polish soccer fans, the authors hypothesize and test the relationships of various social influence processes and perceived corporate social responsibility on desire and on we-intention to use social networking systems for collection action

The Study for Public Management Policy Utility Evaluation and Optimization System under the Framework of Social Computing Perspective

In recent years, in order to rationalize the allocation of social resources and optimize the implementation of public management policies, scholars have conducted in-depth researches on policy effectiveness. However, at present, most of the study is still at the level of using macrolevel qualitative analysis, and lack of quantitative analysis and evaluation system for the effectiveness of policy implementation. The goal of this article is to discuss the utility evaluation system of public management policy from the perspective of social computing. First, based on the data obtained through questionnaire survey, we obtain indicators of the survey data by using factor analysis, and a new BDI (belief–desire–intention) model is created based on the observation indicators, and then the simulation platform is constructed; then, a brand new quantitative analysis method for policy optimization is proposed by using modified logistic functions as a tool. As application, we conducted the case study for the “Targeted poverty alleviation policy in Yulin region” (Guangxi, China), in which the key indicators for the poverty were established, and then the policy optimization suggestions were given based on the results of simulation experiments. This case study has Chinese characteristics, which might be applied to the poverty alleviation work globally.

A belief–desire–intention model for blog users’ negative emotional norm compliance

PurposeThe purpose of this paper is to understand blog users’ negative emotional norm compliance decision-making in crises (blog users’-NNDC).Design/methodology/approachA belief–desire–intention (BDI) model to evaluate the blog users’-NNDC (the BDI-NNDC model) was developed. This model was based on three social characteristics: self-interests, expectations and emotions. An experimental study was conducted to evaluate the efficiency of the BDI-NNDC model by using data retrieved from a popular Chinese social network called “Sina Weibo” about three major crises.FindingsThe BDI-NNDC model strongly predicted the Blog users’-NNDC. The predictions were as follows: a self-interested blog user posted content that was targeting his own interests; a blogger with high expectations wrote and commented emotionally negative blogs on the condition that the numbers of negative posts increased, while he ignored the norm when there was relatively less negative emotional news; and an emotional blog user obeyed the norm based on the emotional intentions of the blogosphere in most of the cases.Research limitations/implicationsThe BDI-NNDC model can explain the diffusion of negative emotions by blog users during crises, and this paper shows a way to bridge the social norm modelling and the research of blog users’ activity and behaviour characteristics in the context of “real life” crises. However, the criterion for differentiating blog users according to social characteristics needs to be further revised, as the generalizability of the results is limited by the number of cases selected in this study.Practical implicationsThe current method could be applied to predict emotional trends of blog users who have different social characteristics and it could support government agencies to build strategic responses to crises.Originality/valueThis paper supports the creation of normative models and engineering methods to predict blog users’-NNDC and mitigate their effect in real-world crises.

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.

A structural model for destination travel intention as a media exposure

Purpose This study aims to investigate the effects of a prospective traveler’s perception of media exposure on their intention to visit a destination (i.e. South Korea). Cultural exposure to a particular country through media affects people’s preference for that foreign country, and may ultimately be a function of the behavior for consuming that country’s cultural products – e.g. traveling to that country. Media exposure has been recognized as a major underlying reason for the desire to visit a destination. Design/methodology/approach This study examines the impacts of potential travelers’ media exposure in three different language-use groups (i.e. English, Japanese and Chinese) and their perception of the media exposure on their intention to visit the actual site (i.e. South Korea). To enhance the understanding of the intention to visit the destination, this study proposes a research model based on use and gratification theory and the belief–desire–intention model. Findings Mass and social media exposure had an effect on the intention to visit a destination as a result of the gratification and desire experienced through the content. Research limitations/implications This study suggests the synthesis of the use and gratification theory and the belief–desire–intention model and an examination of theoretical and practical implications. Originality/value This study involved a sample of users of destination marketing sites. In addition, this study investigated the users’ intentions to visit a real tourism destination taking into consideration mass media (traditional media) and social media (new media) based on the use of gratification theory and the belief–desire–intention model. Practically, the findings highlight the crucial role of social media in the intention to visit the tourism destination.

Available Bandwidth Estimation Using Collision Probability, Idle Period Synchronization and Random Waiting Time in MANETs: Cognitive Agent Based Approach

In Mobile Ad hoc NETworks (MANETs), each node estimates available bandwidth (AB) before transmission of real-time multimedia data, as the channel is shared. The existing AB estimation techniques lack accuracy and add overhead. Accuracy of these AB estimation techniques could be enhanced by adopting some intelligent techniques. Intelligent software agents such as cognitive agent (CA) have significant potential to solve the challenges of estimating AB in MANETs. Intelligence is provided similar to the logical thinking like human for decision making in CA. The predominant CA architecture is belief desire intention model, which performs the task on behalf of human user as an assistant to him/her. In this paper, we propose CA-based available bandwidth estimation using collision probability, Idle period synchronization and random waiting time (BECIT) in MANETs. The proposed BECIT technique uses CA at each node to create mobile agents for collecting and distributing the statistics (such as idle period, packet loss, and AB, etc.) over the network. The collected network statistics are used by CA to estimate AB using our previous work distributed lagrange interpolation based AB estimation (DLI-ABE). The DLI-ABE comprises of statistical models for idle period synchronization by differentiating their states as BUSY, IDLE, and SENSE BUSY; collision probability using distributed Lagrange Interpolation polynomial; and random waiting time that consumes AB such as request-to-send, clear-to-send commands in IEEE 802.11 standard. Estimation of end-to-end AB on a multi-hop path uses each node AB. The simulation results show that BECIT performs approximately 30 % more accurate than the non-agent based AB estimation scheme. Memory overhead is essential to store knowledge base for agents whereas the computational overhead is due to Lagrange Interpolation polynomial at each node in the proposed scheme.

Improving cognitive agent decision making: Experience trajectories as plans

In task environments with large state and action spaces, the use of temporal and state abstraction can potentially improve the decision making performance of agents. However, existing approaches within a reinforcement learning framework typically identify possible subgoal states and instantly learn stochastic subpolicies to reach them from other states. In these circumstances, exploration of the reinforcement learner is unfavorably biased towards local behavior around these subgoals; temporal abstractions are not exploited to reduce required deliberation; and the benefit of employing temporal abstractions is conflated with the benefit of additional learning done to define subpolicies. In this paper, we consider a cognitive agent architecture that allows for the extraction and reuse of temporal abstractions in the form of experience trajectories from a bottom-level reinforcement learning module and a top-level module based on the BDI (Belief-Desire-Intention) model. Here, the reuse of trajectories depends on the situation in which their recording was started. We investigate the efficacy of our approach using two well-known domains – the pursuit and the taxi domains. Detailed simulation experiments demonstrate that the use of experience trajectories as plans acquired at runtime can reduce the amount of decision making without significantly affecting asymptotic performance. The combination of temporal and state abstraction leads to improved performance during the initial learning of the reinforcement learner. Our approach can significantly reduce the number of deliberations required.

Group planning with time constraints

Embedding planning systems in real-world domains has led to the necessity of Distributed Continual Planning (DCP) systems where planning activities are distributed across multiple agents and plan generation may occur concurrently with plan execution. A key challenge in DCP systems is how to coordinate activities for a group of planning agents. This problem is compounded when these agents are situated in a real-world dynamic domain where the agents often encounter differing, incomplete, and possibly inconsistent views of their environment. To date, DCP systems have only focused on cases where agents’ behavior is designed to optimize a global plan. In contrast, this paper presents a temporal reasoning mechanism for self-interested planning agents. To do so, we model agents’ behavior based on the Belief-Desire-Intention (BDI) theoretical model of cooperation, while modeling dynamic joint plans with group time constraints through creating hierarchical abstraction plans integrated with temporal constraints network. The contribution of this paper is threefold: (i) the BDI model specifies a behavior for self interested agents working in a group, permitting an individual agent to schedule its activities in an autonomous fashion, while taking into consideration temporal constraints of its group members; (ii) abstract plans allow the group to plan a joint action without explicitly describing all possible states in advance, making it possible to reduce the number of states which need to be considered in a BDI-based approach; and (iii) a temporal constraints network enables each agent to reason by itself about the best time for scheduling activities, making it possible to reduce coordination messages among a group. The mechanism ensures temporal consistency of a cooperative plan, enables the interleaving of planning and execution at both individual and group levels. We report on how the mechanism was implemented within a commercial training and simulation application, and present empirical evidence of its effectiveness in real-life scenarios and in reducing communication to coordinate group members’ activities.

Model checking agent programming languages

In this paper we describe a verification system for multi-agent programs. This is the first comprehensive approach to the verification of programs developed using programming languages based on the BDI (belief-desire-intention) model of agency. In particular, we have developed a specific layer of abstraction, sitting between the underlying verification system and the agent programming language, that maps the semantics of agent programs into the relevant model-checking framework. Crucially, this abstraction layer is both flexible and extensible; not only can a variety of different agent programming languages be implemented and verified, but even heterogeneous multi-agent programs can be captured semantically. In addition to describing this layer, and the semantic mapping inherent within it, we describe how the underlying model-checker is driven and how agent properties are checked. We also present several examples showing how the system can be used. As this is the first system of its kind, it is relatively slow, so we also indicate further work that needs to be tackled to improve performance.

Towards intelligent agent based software for building related decision support

To automatically deliver enhanced knowledge relating to building use, for tasks such as facility management (FM), a software system has been developed that exploits the combination of a number of technologies. As well as generating useful knowledge for decision support, the software aims to remove reliance on expert users, be self configuring, continually adapt to the environment, and employ learning to evolve its performance. The system is realised centrally by a multi agent society in which the agents are characterised by the strong notion of agency using the BDI (belief, desire, intention) model. In most existing applications in construction agents are task focussed rather than the goal focussed deliberative agents employed here. The BDI model is a ‘natural’ (human) abstraction for modelling complex systems, and goals are a stable way to define required behaviour. The agents are supported by a range of ontologies describing the semantics of the domain as well as aspects of agents’ goals. Furthermore the agents utilise a distributed network of readily available wired and wireless sensors and associated data storage providing access to near real time and historical data, as well as an Industry Foundation Classes (IFC) model describing building geometry and construction. The system produced can be used by non-specialist users, simply requiring an IFC specification of the building and sensor locations. Agents infer the roles of the sensors from an ontology together with context, and assign appropriate roles. The agents individually and cooperatively work towards identifying the usage and dynamics of arbitrarily sized spaces in buildings. Such knowledge can be used to support FM decisions such as the optimisation of the energy consumption/environmental comfort demand trade-off. Negotiation is used to increase robustness as well as to fill in missing information. The limitations of practical application of the technologies that failed to deliver expected benefit are also detailed.

A Multi-agent Architecture Based on the BDI Model for Data Fusion in Visual Sensor Networks

The newest surveillance applications is attempting more complex tasks such as the analysis of the behavior of individuals and crowds. These complex tasks may use a distributed visual sensor network in order to gain coverage and exploit the inherent redundancy of the overlapped field of views. This article, presents a Multi-agent architecture based on the Belief-Desire-Intention (BDI) model for processing the information and fusing the data in a distributed visual sensor network. Instead of exchanging raw images between the agents involved in the visual network, local signal processing is performed and only the key observed features are shared. After a registration or calibration phase, the proposed architecture performs tracking, data fusion and coordination. Using the proposed Multi-agent architecture, we focus on the means of fusing the estimated positions on the ground plane from different agents which are applied to the same object. This fusion process is used for two different purposes: (1) to obtain a continuity in the tracking along the field of view of the cameras involved in the distributed network, (2) to improve the quality of the tracking by means of data fusion techniques, and by discarding non reliable sensors. Experimental results on two different scenarios show that the designed architecture can successfully track an object even when occlusions or sensor’s errors take place. The sensor’s errors are reduced by exploiting the inherent redundancy of a visual sensor network with overlapped field of views.

THE UNEXPECTED ASPECTS OF SURPRISE

Some symbolic AI models, for example, BDI (belief, desire, intention) models are conceived as explicit and operational models of the intentional pursuit and belief dynamics. The main concern of these models is to provide a clear understanding of the functional roles of different kinds of epistemic and motivational states (beliefs, acceptances, expectations, intentions, goals, desires, etc. …), of the relational properties among them. Mental configurations of appraisal (involving different kinds of motivational and epistemic states) which correspond to particular cognitive emotions such as disappointment, fear, relief, shame, etc. … have been analyzed by several authors close to the BDI theoretical tradition. The main objective of this work is a formal analysis of Surprise in a BDI-like cognitive architecture. A clarification of the functional role of Surprise in a BDI-like cognitive architecture with respect to resource bounded belief revision is given.

Decision support for ethical problem solving: A multi-agent approach

This paper suggests that a multi-agent based decision aid can help individuals and groups consider ethical perspectives in the performance of their tasks. Normative and descriptive theories of ethical problem solving are reviewed. Normative theories are postulated as criteria used with practical reasoning during the problem solving process. Four decision aid roles are proposed: advisor, group facilitator, interaction coach, and forecaster. The research describes how the Theory of Planned Behavior from psychology can inform agent processing. The Belief–Desire–Intention model from artificial intelligence is discussed as a method to support user interaction, simulate problem solving, and predict future outcomes.