Engineering cooperative smart things based on embodied cognition

Abstract

Abstract-The goal of the Internet of Things (IoT) is to transform any thing around us, such as a trash can or a street light, into a smart thing. A smart thing has the ability of sensing, processing, communicating and/or actuating. In order to achieve the goal of a smart IoT application, such as minimizing waste transportation costs or reducing energy consumption, the smart things in the application scenario must cooperate with each other without a centralized control. Inspired by known approaches to design swarm of cooperative and autonomous robots, we modeled our smart things based on the embodied cognition concept. Each smart thing is a physical agent with a body composed of a microcontroller, sensors and actuators, and a brain that is represented by an artificial neural network. This type of agent is commonly called an embodied agent. The behavior of these embodied agents is autonomously configured through an evolutionary algorithm that is triggered according to the application performance. To illustrate, we have designed three homogeneous prototypes for smart street lights based on an evolved network. This application has shown that the proposed approach results in a feasible way of modeling decentralized smart things with self-developed and cooperative capabilities.

Artificial Intelligence versus Software Engineers: An Evidence-Based Assessment Focusing on Non-Functional Requirements

“…Subsequently, we compared the solution produced by the learning algorithm with those proposed by the engineers. For this particular application, a 'more appropriate' solution is defined as one that either outperforms others in the main scenario 13 , or is reusable in a new context, guided by factors such as power consumption and safety-referring to maximum illumination in lit areas.…”

“…Subsequent research has delved into the application of artificial intelligence, predominantly machine learning (ML) techniques, for automating diverse software engineering (SE) tasks 3 . These explorations encompass methods for i) project management, addressing issues related to cost, time, quality prediction, and resource allocation 4 ; ii) defect prediction 5 ; iii) requirements engineering, concentrating on classification or representation of requirements 6,7 , or generation of requirements 8 ; iv) software development tasks, such as code creation [9][10][11][12] , synthesis 13 , and code assessment 14 ; v) testing, like detecting and fixing compilation or build errors 15,16 ; vi) software maintenance tasks, such as renaming software entities (e.g., variables, methods) with meaningful identifiers 17 ; and vii) data analysis, for instance, tackling data science issues 18 .…”

Towards an Agent-Based Architecture using Deep Reinforcement Learning for Intelligent Internet of Things Applications.pdf

“…It is very hard, however, to completely define a physical system's behaviors at the time of design and to recognize and promote characteristics that contribute to attractive collective behavior. In order to solve these issues, some approaches ( [12,13]) have suggested using evolving machine learning such as neural networks to enable an embodied agent to learn how to adapt their behavior in a dynamic environment. In fact, through learning skills, agents will be able to reason conveniently, create a suitable policy and make good decisions [5].…”

Governance in Adaptive Normative Multiagent Systems for the Internet of Smart Things: Challenges and Future Directions

“…Several proposals have been provided in each of the areas, namely governance systems and smart things, namely investigations on areas related to governance research [3] [1] [12] [5] and on smart things based on multi-agent systems [8] [22] [13]. However, each of the respective research communities have worked in isolation, and we still lack a common and integrated solution that supports the integration of governance oriented adaptive normative MAS and smart things.…”

“…To deal with this challenge it is necessary to add to the metamodeling language proposed in [19] the concepts of coordination, organization, institution to support governance features. Without modeling these new features, it will not be possible to represent certain classes of relevant and naturally solvable problems by SMAs, such as those related to autonomous vehicles [14], smart lights [8] [17], smart cities [6]. In contrast, modeling these abstractions would make it possible to represent every smart thing that participates in an open environment as part of a governance oriented adaptive normative MAS, with sensors, actuators and algorithms capable of (autonomous) decision making as the environment undergoes changes.…”