Figure - available from: Journal of Ambient Intelligence and Humanized Computing
This content is subject to copyright. Terms and conditions apply.
Source publication
Mobile sensing systems based on smartphones, connected vehicles and integrated sensors on new mobile devices have become an important alternative for the development of intelligent services in large urban environments. Massive data collection and its real-time analysis are essential for big cities to move towards energy efficiency, sustainable mobi...
Citations
... In this sense, various contributions were proposed to deal with community detection in such systems. However, considering the importance of multi-agent modeling systems for the coordination and/or cooperation processes, the MAS represents a great potential to contribute in modeling the community detection problem, into autonomous and intelligent agents concerning environmental characteristics [14,22,26]. ...
In recent years, community detection has emerged as an important field of research, exerting a profound influence on various domains such as Social networks, Recommender systems, Citation networks, and Enterprise network. Acknowledging the profound implications of this development, we introduce a novel approach integrating a data mining technique, leveraging topical attributes of a network’s components. This approach seamlessly integrates with Social Network Analysis, within a multi-agent architecture composing four distinct hierarchical levels. In this paper, we present a novel approach for community detection that redirects the focus from traditional topological properties to topical properties of nodes and edges. This topical analysis perspective, often-neglected, constitutes the core of the proposed three-step methodology. We leverage the power of association rule mining using the Apriori algorithm as the initial step, extracting valuable insights from the network. Subsequently, we meticulously select meaningful rules, preparing them for the final stage where the proposed algorithm execution identifies both overlapped and non-overlapped communities within the network. To evaluate the effectiveness of our multi-agent system approach, we conducted tests on several real-world social networks, and performed comparisons with six traditional methods, thereby confirming the foundations of our approach.
... When equipped with sensors, multiple vehicles can autonomously navigate to different locations to collect distributed environmental data. This paradigm, often referred to as mobile sensing, has attracted attention from a variety of disciplines, such as air quality sensing [1], traffic monitoring [2], fire detection [3], etc. For example, in a smart home, multiple devices (e.g., sweeping robots) can cooperate to sense the environment and perform related tasks [4], such as cleaning and tidying. ...
In a typical mobile-sensing scenario, multiple autonomous vehicles cooperatively navigate to maximize the spatial–temporal coverage of the environment. However, as each vehicle can only make decentralized navigation decisions based on limited local observations, it is still a critical challenge to coordinate the vehicles for cooperation in an open, dynamic environment. In this paper, we propose a novel framework that incorporates consensual communication in multi-agent reinforcement learning for cooperative mobile sensing. At each step, the vehicles first learn to communicate with each other, and then, based on the received messages from others, navigate. Through communication, the decentralized vehicles can share information to break through the dilemma of local observation. Moreover, we utilize mutual information as a regularizer to promote consensus among the vehicles. The mutual information can enforce positive correlation between the navigation policy and the communication message, and therefore implicitly coordinate the decentralized policies. The convergence of this regularized algorithm can be proved theoretically under certain mild assumptions. In the experiments, we show that our algorithm is scalable and can converge very fast during training phase. It also outperforms other baselines significantly in the execution phase. The results validate that consensual communication plays very important role in coordinating the behaviors of decentralized vehicles.
... In [23], an architecture based on a MAS and big data analysis was proposed to exploit, analyze and process the data accumulated by sensors in a smart city, including those provided by connected vehicles and/or those equipped with integrated sensors. MASs have been also integrated into the architecture of real-time autonomous vehicle infrastructure simulation plat-forms [24,25] and used to evaluate the impact of V2V and V2I technologies on the mobility performance [26]. ...
Road traffic is drastically increasing in big cities around the world. Inorder to enable a flexible management of this traffic, Intelligent TransportationSystem (ITS) solutions are relying on emergent ubiquitous, mobile, and communi-cation technologies, particularly to intelligently deal with the limited capacities ofthe existing road infrastructures. While intelligence is left to the autonomous andconnected vehicles as well as to the ITS, the road infrastructure has been mostlyplaying a passive role (as a source of data). Road signage, in particular, are in bestcases dynamic but do not play an active role in monitoring traffic and incidents.We propose in this paper to build Smart Road Signs (SRS) that can collaboratewith Connected Vehicles in order to monitor traffic and warn drivers about any in-cident or danger. Our SRSs are meant to operate autonomously in order to detectroad traffic problems, share appropriate information with vehicles in the vicinity,and display relevant messages based on the ongoing contextual situation. To meetour goals, we rely on Multi-Agent Systems to design SRSs as proactive componentsin the ITS landscape. We also rely on agent mobility in order to strengthen thecollaboration with the connected vehicles.
... Wang et al. [12] proposed a multi-agent system based on generalized covariance intersection for multi-view surveillance in centralized and decentralized situations. Laport et al. [13] proposed a multiagent architecture for collecting massive data using mobile sensing devices. Jing et al. [14] introduced a coverage path planning framework for the large and complex structure inspection of multiple unmanned aerial vehicles. ...
As the roles of robots continue to expand in general, there is an increasing demand for research on automated task planning for a multi-agent system that can independently execute tasks in a wide and dynamic environment. This study introduces a plugin framework in which multiple robots can be involved in task planning in a broad range of areas by combining symbolic and connectionist approaches. The symbolic approach for understanding and learning human knowledge is useful for task planning in a wide and static environment. The network-based connectionist approach has the advantage of being able to respond to an ever-changing dynamic environment. A planning domain definition language-based planning algorithm, which is a symbolic approach, and the cooperative–competitive reinforcement learning algorithm, which is a connectionist approach, were utilized in this study. The proposed architecture is verified through a simulation. It is also verified through an experiment using 10 unmanned surface vehicles that the given tasks were successfully executed in a wide and dynamic environment.
... For example, weak learner models using local data allow getting local decision-making models. Thus, immediate recommendations are possible, providing quick answers in fire detection, flood detection, water monitoring, among too many other situations [7]. ...
The Internet of Things has been raised as an alternative for implementing different data collection strategies. Because of the limited hardware, it is complemented with edge computing trying to bring the computing power as close to the data source as possible. Collected data have particular importance for building different kinds of artificial intelligence models. However, it requires that data are transported from the data source to some centralized processing place (e.g., cloud). Data ownership and data privacy associated emerge as essential concerns. Thus, federated learning is proposed as an alternative to train models in a distributed way and avoid unnecessary data transportation. So, a global model is built based on local parameters/weights computed in distributed nodes. This work describes a Systematic Mapping Study focused on recent applications of federated learning on the Internet of Things and Edge environments. Study cases and applications from 2021 onwards are queried and analyzed from the Scopus database. The study is limited to articles written in the English language and published in journals. From 32 articles, 28 are retained following the inclusion and exclusion criteria. Data Ownership emerges as the main challenge, while the model performance represents the main trend. China and United States concentrate 55 % of the involved articles on the subject.
... Zhang (2019) introduced a quantum particle swarm optimization (QPSO) algorithm to support task allocation in multi-agent systems (MAS). Laport et al. (2019) presented a MAS system using domain expert learning agents to support data sensing in different domains. Some methods in literature employ ontologies to support agents in performing cooperatively their tasks (Schmidt et al. 2015;Okresȃ Durić et al. 2018). ...
In recent years, the role of unmanned vehicles (UVs) is increased in many surveillance applications; they are substituting the humans in many risky activities, especially when cooperative tasks from UV team are required. To this purpose, this paper presents an agent-based framework that models a multi-UV system for surveillance applications. The agents act as wrappers for the different types of UVs, that capture data from the scene (in the area of the UV mission) and then process them, each one according to its own skills and features. The collected and processed data are then shared from the agent team to find a common agreement on the comprehension and criticality assessment of the scenario. The agent paradigm provides a seamless framework for UV interaction, making the different methodologies and technologies, designed for the different UV types, transparent. The proposal shows the agent-based modeling for a multi-UV system, where each agent hides the facilities and features of the UV it wrapped, with the aim of deploying a homogeneous interface to facilitate the collective scenario assessment in terms of critical or alerting issues, detected in the evolving scene.
... MAS agents can parcel out or perform tasks with other artificial agents or even with humans. These systems have demonstrated its relevance in several realistic domains, namely the mobile sensing systems in large cities (Laport et al. 2019), decision support systems (Amato et al. 2014) and the minimization of traffic among the road network (Krishnan and Ram 2018). ...
Multi-agent systems (MAS) consist of autonomous agents that parcel out different tasks and make decisions in dynamic environments. Distributed constraint satisfaction problem (DisCSP) is the most effective and applicative MAS framework. In DisCSP, each agent is connected to the other agents via constraints and holds its own local constrained problem. Those agents find solutions satisfying their own constraints and the linking ones too by collaborating and exchanging messages holding their instantiations. This formalism does not take into consideration the possibility of the presence of unethical agents which can make irrelevant or even dangerous decisions, especially when human agents are involved in the resolution. In this paper, we propose an extension of the DisCSP into an ethical formalism “E-DisCSP”. It allows to control agents, detect intrusions and apply the convenient actions when an unethical agent is picked up. All these functionalities are done via control framework, in order to maintain the DisCSP resolution as normal as possible. Experimental results show the efficiency of our contribution. The detection rate of unethical agents achieves up to 100%. And the convenient actions’ application allows, to go from 45 to 0% of wrong solutions.
