The evolution of manufacturing, driven by the Fourth Industrial Revolution, demands adaptable and interconnected production methodologies to meet the increasing demand from consumers for personalised products. Cyber- Physical Production Systems have emerged as a crucial innovation in this context, facilitating virtual representations of production resources and enabling dynamic interactions between computational and physical processes. This article proposes a framework integrating Cyber-Physical Production Systems with Digital Twins to understand manufacturing systems and enhance their functionality comprehensively. The framework comprises three layers: Physical Assets, Modular Cyber-Physical Production System, and Digital Twin. Utilising Multi-Agent systems, the Modular Physical Production System orchestrates production processes, optimises efficiency, and ensures adaptability to market fluctuations. Meanwhile, the Digital Twin serves as a real-time representation of the physical system, facilitating monitoring, interaction, and understanding of the system's behaviour. The implementation involves creating 3D models using Blender and a virtual environment using Unity. The proposed framework has demonstrated its effectiveness in addressing the complexities of modern manufacturing, fostering adaptability, and facilitating collaboration between humans and machines.

Scientists play a pivotal role in advancing our understanding of the world, contributing not only through objective and rational work but also by incorporating their emotions into scientific discourse. The affective side of science is gaining increased attention. Building on this line of inquiry, this study explores the impact of emotional elements and sentiment analysis within abstracts published in the journal Nature over the past 10 years, adopting an affective-centric perspective on scientific endeavors. The research aims to explore the presence and trends of emotional elements and sentiments in these abstracts and to investigate the correlation between these emotional elements and citation counts, considering the influence of emotions on decision-making. The novelty of this research lies in not only considering sentiment analysis but also employing two distinct emotion detection techniques on the abstracts. The study employs a hybrid approach that integrates Lexicon-based, Deep Learning (DL), and Natural Language Toolkit (NLTK) sentiment analysis methods to explore diverse emotional dimensions through text emotion detection. The results demonstrate the presence of emotions and positive sentiment trends in the abstracts and positive sentiments in scientific abstracts were more prevalent and are associated with higher citation rates. It also highlights the observation that diverse emotions are prevalent in the most cited papers.

Self-configuration in manufacturing is a key trend to generate adaptable production systems. Different product requirements need different machine settings and continuous software update. Existing approaches usually assume that manufacturing resources (e.g. robotic platforms) have already a predefined set of available capabilities or that there is a centralized manager able to provide configuration updates. Centralized approaches are not always a suitable solution. Dynamic changes in production require continuous maintainability of the central server, constraining the shop-floor agility. New approaches should consider emergent self-configuration i.e. carried out at run time and decoupled from a centralized unit. In this work we present a framework for self-configuration of robotic platforms, where those are not explicitly preconfigured; instead, control parameters are transferred wirelessly from raw material to resources during production (once they arrive to the resource), following an intelligent product-driven manufacturing approach. A set of templates are proposed to generalize the sequential behaviour of manufacturing operations. Thus, manufacturing resources can read, use these parameters and store them for future operations. This framework is illustrated with an assembly operation using an educational robotic platform.

Fourth industrial revolution, also known as Industry 4.0 (I4.0), was suppose to be a transformative revolution that enables industries to move from being a technology and product-centred premise to become a human-centric one. However, the development of this perspective within I4.0 has not been yet fulfilled. In this regard, the idea of Industry 5.0 (I5.0) is focused on such a concept of human-centeredness and implies that technology, organizations, and individual working places must be adapted to human and societal needs. By breaking down relevant and recent literature in a systematic way, present article analysed three specific aspects related to human-centric manufacturing include understanding the role of humans, the required labour skills, and the various challenges that I5.0 has attempted while placing the human perspective as a fundamental industrial pillar. It also delivered a Roadmap for I5.0 skills exploitation and perform outreach activities which contribute to the backbone of I5.0, while having a discussion towards human-cantered, and technology-supported jobs.

In last decades, novel manufacturing requirements have introduced the need of having higher levels of automation in manufacturing shop-floors. Supervisory and centralized approaches are not able to provide the scalability, modularity and autonomy required in the fourth industrial revolution. Self-organizing ideas from nature are very promising and have not been highly explored in manufacturing automation. The contribution of this paper goes towards that direction. We first define a highly flexible shop-floor and the interaction of the elements is proposed considering a multi-agent based negotiation. From this point, we introduce and define analogies and patterns based on the echolocation of bats that can be applied in manufacturing automation. These patterns are based on the way how bats hunt and avoid obstacles based on the emission of sound and echo perception. These ideas are showcased using the matrix production concept which proofs the feasibility of the current approach. Various conclusions, limitations and potential future works are described at the end of the paper.

The advent of Industry 4.0 and the development of future industrial applications can be achieved using Cyber-Physical Systems (CPS). This technological development invokes high levels of communication and computation in the form of an interconnected network of industrial resources. Multi-agent systems precisely can empower such technological evolution by introducing properties like decentralisation, autonomy, flexibility, social ability and modularity to the industrial context. In this regard, the current work surveys recent multi-agent based manufacturing approaches and provides a general vision of current trends focusing on frameworks/architectures, complementary technologies and common applications. This article ends with an integrated discussion of emerging agent-based industrial challenges, a general conclusion and final remarks.

The advent of Industry 4.0 has made it crucial to improve the accessibility of higher education and to ensure that the future generation of engineers is able to acquire interdisciplinary competences to face the challenges of the data-driven era, including hands-on experience with real scenarios. We present an approach for teaching Cyber-Physical Production Systems remotely to graduate students, along with a discussion of sample projects, recommendations and the lessons learned in the effort to break down geographical barriers to education, reduce the cost associated with material and mitigate the impact of unforeseen disruptions such as the one caused by the pandemic scenario of COVID-19 in 2020. A combination of physical learning factory demonstrators, digital twin and simulation scenarios was developed to provide students with the resources to remotely implement an end-to-end Cyber-Physical Production System, along with its integration with other key technologies of Industry 4.0. A case study at the NOVA University of Lisbon showed that average attendance improved by 26.6%, retention in lab component improved by 12.9% and lab grades improved on average by 7.33% compared to on-site iterations of the same course in the two previous years.

This paper presents a platform-agnostic distributed navigation architecture for autonomous mobile robots operating in intra-factory logistics. Communication, control, navigation, diagnosis and hardware are layered in a hierarchical approach increasing robustness, modularity and flexibility. This architecture promotes several key features, such as dynamic selection of navigation profiles, semantic mapping and human-aware navigation. The approach allowed multiple autonomous mobile robots, cooperating through a fleet management system, to adapt to a wide range of situations, alternating their path planning between high-speed free-space strategies, and high precision low-speed for tight passageways and docking to assembly stations. The benefits of the proposed architecture were validated through a set of experiments in a mockup shopfloor environment. During these tests 3 robots operated continuously for several hours, self-charging without any human intervention.

The fundamental role played by “new technologies” to enhance the manufacturing infrastructure, products and services is confirmed by the strategies, roadmaps and initiatives established by the developed countries such as EU-28, US, China and Japan. Putting the light on Europe, the key priorities for manufacturing are aligned with the Industry 4.0 strategy/program where the proliferation of cyber-physical systems (CPSs) and technologies like predictive data analytics, cloud and edge computing are creating the foundation for smart factory, i.e. the efficient and effective connection between products, processes and their related services. However, even if the smart factory vision is quite clear, in practice, it is still unclear how it can be implemented in a way to allow the transparent data exchange between all the layers of a manufacturing company, as well as, between the manufacturing company and the value chain partners. Within the German Industrie4.0 program, the Asset Administration Shell concept is defined to create a standardized digital representation of the asset while ensuring interoperability between all the applications within the manufacturing ecosystem. In this paper, an implementation of the asset administration shell is provided - also called NOVAAS - based on the deep usage of internet technologies, languages and software engineering techniques and methods (such as DevOps, microservices, continuous integration, etc.). The main goal is to contribute towards a generic and extensible reference implementation of the concept.

In the last years the utilization of Multiagent Systems to implement distributed control systems in industrial environments was presented as suitable and as a cost-effective solution to deal with the new requirements regarding flexibility and dynamism on the shop-floor. However, the proposed implementation of these distributed Cyber-Physical Production Systems faced some challenges regarding hardware and network requirements. Hence, the proposed work presents one utilization of a Multiagent-based distributed control system running on the fog level and running upon the edge level. This research presents a test and assessment of running intelligent agents outside the edge level but at the same time avoid the deployment of the industrial agents on the cloud level due to time and performance constraints. The proposed test presents a Multiagent architecture responsible for controlling the shop-floor, but the overall architecture was designed to accommodate the agents on the fog level, running upon the edge level composed by industrial controllers running Device Profile Web Services.

Resilience measurement can be viewed as a multicriteria hierarchical decision making problem since calculating the final level of resilience involves measuring different criteria, at several hierarchical levels, and then merging the information together. In this paper, a resilience model for disaster rescue networks is discussed with a full-reinforcement operator, denoted continuous reinforcement operator. This approach is tested with different levels of reinforcement and the results are compared with those from a Fuzzy Inference System. The proposed approach offers interesting features to support balanced development of disaster rescue networks and facilitates managerial decisions by imposing standards for criteria to penalize or reward the information fusion process.

Cyber-Physical Systems (CPS) are creating new market opportunities and business models for all kind of European Industries. CPS-based platforms are increasing in their size and target application areas in a steady manner. However, even if progress is made every day supported by continuous technological advancements, CPS application and deployment is still challenging. Many solutions have been made available or is currently under development in several research projects/ initiatives. Typically, these solutions show no interoperability between each other and are tailored to a specific application context. Thus, there is an urgent need for a clear definition of what a CPS-populated system actually is. This will provide a common ground for designing and building interoperable CPS-populated systems. Interoperability represents one of the most challenging problems for such systems essentially due to their intrinsic characteristics: heterogeneity, distribution and networked. These must be addressed to allow the cooperation and collaboration between all the actors of the system. In this landscape, the MANTIS project is aimed to provide a reference model for interoperable and interconnected CPS-populated systems for maintenance-related ecosystems, which is the focus of this paper.

In the last decades, several research initiatives suggested new solutions regarding the interoperability and interconnectivity among heterogeneous production components and all the actors that somehow interact within the shop-floor. However, most of the proposed data representations are focused on the description of the production capabilities. In this paper, it is proposed a common data model focused not only in the production capabilities of the different components as well as the description of all the events, variables and resources that could indicate quality issues. Hence, the proposed data model describes all the information required by the GO0DMAN solution to reduce as much as possible, the defects, the respective causes and the strategies to avoid the propagation along the line. In order to increase the adoption of the proposed data model, it was developed using AutomationML. The proposed data model was designed and tested within the scope of the Horizon 2020 GO0DMAN project.

Smart grids rely on the integration of distributed energy resources towards an intelligent and distributed manner to organize the electrical power grid enabled by a bidirectional flow of information to improve reliability and robustness, fault detection and system operation, and plug-and-playability of energy devices. The integration of information and communication technologies (ICT), one of the key enablers of smart grids, will ease the deployment of intelligent and distributed systems implementing the automation functions. In this context, there is a need to assess how these systems, developed using these emergent technologies, e.g., multi-agent systems, data analytics and machine learning, will behave and affect the working conditions of the power grid. This paper aims to explore the development of a transparent and loose-coupled interface between the behavioral control system and the physical or simulated power system environment, in a coupled simulation perspective, aiming to assess and improve the development of such systems during the design phase.

The objective of the work leans on the development of an agent-based engine, part of a Service Oriented Architecture, that provides the mechanisms to receive, interpret and execute extensions of products/processes (as software). A novel methodology that uses Service Composition paradigm as base for development and specification of features used by the acting services, along with the use of specific Engineering Tools dedicated to the parameterization of each service, provides the encapsulation of software solutions which the deployment environment can solve and act in accordance on a real-time deployment environment. This concept has a wide range of applicable areas, however, the analysis of the state of the art reveals that this is an emergent target to Manufacturer Companies, as there are missing effective solutions that provide the viability for these companies to perform engineering of Product Service Systems around their products, enabling their extension as services and providing the construction of new business models. The work is driven by industrial application scenarios which provide a novelty of aspects of product-systems systems in different sectors (automotive, home appliances, automation equipment's, etc.) of manufacturing companies.

Service-Oriented Architectures (SOAs) along with Web Services are now being considered as the de-facto standard implementation to connect several layers of a given business in a fast, secured and decentralized fashion. On the other hand, Internet of Things (IoT) devices are generating more and more data that can be analyzed and processed. Despite impossible to analyze locally, this data could be analyzed using a Service-Oriented Architecture. This work presents a solution, contemplating a SOA, that retrieves data from an unintelligent-turned-IoT device, using Representational State Transfer (REST) services. It then uses the ProSEco platform to process the data. Once processed, the results are presented using a web-based visualization tool.

Several research agendas worldwide are targeting the development of Industrial Cyber-physical Systems as the next generation of intelligent embedded devices with improved interaction capabilities. These devices, and their potential uses, are though to deliver a radical increase in system sustainability, reconfigurability and flexibility which is perceived to be the root of the so called 4thIndustrial Revolution. However such technical systems, at the envisioned revolutionary scale, do not exist just yet and require a convergent and multidisciplinary research and development efforts. The academia curricula are also, albeit slowly, adjusting to the emerging education requirements. The Summer School on Intelligent Agents in Automation is a joint effort from several researchers in core areas of the 4thIndustrial Revolution landscape to close the gap and promote advanced education in this context. This paper describes the implementation of the 2ndedition of the event as well as the experience and reflections resultant from it.

The proposed telemetry system consists of a graphical user interface with reconfigurable multiple widgets whose transmission is prioritised based on the user's gaze. The proposed approach encompasses a novel framework for telemetry of remote machines, based on ROS (Robot Operating System). It includes a modular ensemble of GUI, gaze device interoperability, and a ROS sensory topic modulator, which alternates different strategies to optimise all displayed information transmission quality, in a way that best suits the user. The approach was validated on a teleoperation scenario having multiple users control a robot in a designed course.

Cyber-physical systems constitutes a framework to develop intelligent, distributed, resilient, collaborative and cooperative systems, promoting the fusion of computational entities and physical devices. Agent technology plays a crucial role to develop this kind of systems by offering a decentralized, distributed, modular, robust and reconfigurable control structure. This paper describes the implementation of a summer school aiming to enhance the participants' knowledge in the field of multi-agent systems applied to industrial environments, being able to gain the necessary theoretical and practical skills to develop real industrial agent based applications. This is accomplished in an international framework where individual knowledge and experiences are shared in a complementary level.

The world is assisting to the fourth industrial revolution, with several domains of science and technology being strongly developed and, specially, being integrated with each other, allowing to build evolvable complex systems. Data digitization, big-data analysis, distributed control, Industrial Internet of Things, Cyber-Physical Systems and self-organization, amongst others, are playing an important role in this journey. This paper considers the best practices from previous successful European projects addressing distributed control systems to develop an innovative architecture that can be industrially deployed. For this purpose, a particular design process has to be addressed in order to consider the requirements and functionalities from various use cases. To investigate the known practices, four use cases are enlighted in this paper, which cover a wide spectrum of the European industrial force, as well as industrial standards to support a smooth migration from traditional systems to the emergent distributed systems.

With the emergence of the Industry 4.0 concept, or the fourth industrial revolution, the industry is bearing witness to the appearance of more and more complex systems, often requiring the integration of various new heterogeneous, modular and intelligent elements with pre-existing legacy devices. This challenge of interoperability is one of the main concerns taken into account when designing such systems-of-systems, commonly requiring the use of standard interfaces to ensure this seamless integration. To aid in tackling this challenge, a common format for data exchange should be adopted. Thus, a study to select the foundations for the development of such a format is hereby presented, taking into account the specific needs of four different use cases representing varied key European industry sectors.

As web related technologies continue to find their way into production related activities, their true potential can only be explored with a shop floor that is cyber-physically agile. This means the ability to quickly reconfigure a production system during production changeovers while significantly reducing setup times and non value-adding activities keeping the overall performance of the control system. Currently, system reconfiguration in mainly supported by reprogramming. However, reprogramming of large complex systems is often a time consuming and risky activity since it has a non negligible potential for introducing errors which normally result in long debugging times. In this context, this paper details an agent-based system architecture that enables quick system reconfiguration and minimizes the reprogramming effort. For this purpose a production system is abstracted as a multiagent system that dynamically interacts with a native, programmable-logic-controller-based control system and is able to re-parametrize it. The focus on re-parametrization means that the native system can still operate in the absence of the agent system which improves the overall robustness of the solution, using readily available equipment and enables its integration with legacy systems.

This paper presents an incremental learning mechanism for context-aware switching between localisation methods which are available to the robots control system (e.g., GPS-based, map-based). The goal is to avoid the cumbersome and error prone manual mapping between localisation methods and environmental contexts. At each moment, the system determines which localisation method is performing best by comparison with the motion estimates produced by an odometer, assumed as accurate in the short-time. Then, the best performing method is associated to the current environmental context, which is defined by a novel descriptor built from the local occupancy grid. The result of this instance-based learning process is used online to estimate which localisation method performs the best in the current environmental context. The switching process is facilitated by the use of the de facto standard Robot Operating System (ROS) framework. The system was instantiated in a differential-wheeled robot equipped with a short-range 2-D laser scanner, and successfully validated on a set of field trials.

The trends of manufacturing have now begun to call for a more adaptive and dynamic shop floor in the automation industry. The advancements in mobile robotics of recent decades have inspired and cemented a belief that multiple autonomous mobile robotic agents, often cooperatively sharing the workspace with humans, will significantly contribute to a truly flexible shop floor of the future. This article reports on the design specifications for one such robotic agent, the Intra-factory Mobile Assistant Unit (IMAU), able to carry material boxes between supermarkets and assembly stations, dynamically avoiding obstacles, throughout the shop floor. The solution will make shop floor logistics more flexible, reduce the allocated space for buffers, and, also, ease the human workload. The article will cover the sensing, actuation, communication, software architecture, and the seamless integration into the plant's execution system, which shape the autonomous logistics robot to be deployed into a working automotive shop floor during the ROBO-PARTNER project.

In recent years a set of production paradigms were proposed in order to allow companies the possibility to face the new market requirements and needs, such as the new demand for highly customized products and the decrease of product's life cycle. These new paradigms propose solutions capable of facing these requirements, giving manufacturing systems high capacity to adapt along with high flexibility and intelligence in order to deal with disturbances, like unexpected orders or malfunctions. Evolvable Production Systems propose a solution based on the usage of modularity and self-organization to support pluggability and in this way allow the companies to add and / or remove components during execution without any programming effort and without wasting much valued time. Although these new systems give companies the capacity to be more flexible, they were not designed to perform monitoring as flexible as the control system itself, because the usual monitoring software, mostly based on SCADA systems, is not capable of re-adapting during execution to this new plugging and / or unplugging of devices and to changes in the characteristics of the entire system's topology. Considering these aspects, this article proposes a fully distributed agent based architecture, capable of performing monitoring at different levels while still supporting the addition and removal of monitoring entities, responsible for data extraction and analysis during runtime.

Simulation has played an important role along the years to predict systems' behaviour before their deployment. In the case of self-organising mechatronic systems simulation tools can help researchers and practitioners understanding the full potential of the solution as well as its underlying limitations. Self-organising mechatronic systems have passed a feasibility study and presented promising results. However they are rarely explored in industry in part due to the lack of methods to support their design and configuration and the difficulty to predict the systems' behaviour before their deployment. Given the cost and development time associated with building self-organising mechatronic systems this research problem has been left quite unattended. In this article we present a tool that enables the creation and simulation of Evolvable Production Systems and their self-organising behaviour. The generated operational results can posteriorly be used to analyse the suitability of different design and configuration alternatives for different product types and volumes.

Cloud manufacturing (CMfg) arises as a novel paradigm for enabling future manufacturing companies to be responsive, (re)-configurable, adaptable, and flexible, in one word: agile. CMfg allows companies to deploy and manage all the manufacturing information over the network improving enterprises visibility/transparency while opening the doors to the implementation of the global manufacturing enterprise. Thus, CMfg is radically changing the role played by enterprises Information and Communication Technologies (ICT) solutions in both strategic and operational level. In this scenario, agility is a fundamental indicator that has been widely used by academic and industrial communities for evaluating the performance of the network based business such as Supply Chains. It is believed that CMfg can directly affect the agility of an enterprise. Consequently, the main purpose of this work is to analyse the impact of CMfg on Supply Chain agility. Since Supply Chain agility can be categorized in two clusters, namely: capabilities and enablers, then the main objective is to quantify the impact of CMfg on these two categories of agility. To handle the complexity of the problem the DEMATEL model as an effective method for collecting and structuring experts' knowledge is employed. The result could facilitate the investment decision and provide a better understanding on how to use CMfg efficiently to develop Supply Chain Agility (SCA).

Current trends are showing the growing demand of embedded systems, related to cyber-physical systems, which consist of smart physical actors and/or sensors that are reachable e.g. with service-oriented technologies, cooperating and negotiating to behave showing systems-of-systems characteristics and without any defined hierarchical structures. Possible evolvable structures and emergent behaviors allow generating new features/functions by the collaboration of different components of the cyber-physical systems. These Systems-of-Systems characteristics of service-based cyber-physical systems used among others for manufacturing processes are very suitable to support the product life-cycle in the initial (production) phase of a product and are very suitable to increase the energy efficiency of a whole manufacturing system using context sensitive technologies during the running life cycle phase. This paper presents a brief state-of-the-art analysis for service-based manufacturing systems related to product life-cycle support in the initial phase and energy efficiency aspects.

This article describes the first version of a tool designed to infer the network characteristics of JADE-based multiagent systems. The rationale behind the tool is that systems in general and multiagent system in particular, often have some hidden dynamics that contribute to the emergence of desired and undesired characteristics. Traditional sniffing tools simply display the message exchange. The presented tool goes therefore beyond simple message sniffing and infers the agents' network based on the ongoing interactions and codifies it a format suitable for further processing in specialized network analysis tools. In particular the prosped tool identifies the most frequently used communication links and the messages associated with them. To demonstrate the behavior of the tool an exploratory system based on the Evolvable Production System paradigm is discussed and analyzed.

The new market trends are very different, so it is crucial to the companies improve the tools and capabilities that allow themselves readjust rapidly and effectively to the news market changes and to the new requirements. In order to facilitate this process, it is proposed in this paper an agent based implementation that can provide to the existent systems the capacity to quickly adapt and reconfigure using standard technology. The proposed framework provides an intelligent tool to autonomously help the configuration when a production operator pretends to introduce a new variant of product in runtime and consult important information provided by the system to monitor execution.

Computational and communication capabilities are increasingly being used in all devices. In the production context this leads to the generation of massive amounts of data that are rarely proficuously used. More particularly the application of data-mining techniques to infer knowledge from systems' operation to improve its design decisions remains fairly unexplored. This article presents an approach to extract system design and configuration rules from Evolvable Production Systems. Furthermore it provides the empirical results from two test-cases that support the hypothesis that a simulation-data-mining approach can help reducing the complexity of the work carried by system designers and production managers.

The emergence of modern manufacturing paradigms together with the growing interest on distributed architectures has been increasing the use of biologically inspired solutions. However, somehow along the way, developed approaches have converged towards more traditional systems where the physical and logical decoupled nature of the system has been partially lost. In this context, the presented work aims to introduce and analyse a new fully physically and logically decoupled bio-inspired self-organising approach that tries to bring to the mechatronic-agent based manufacturing architectures the dynamics of biological systems. Furthermore, the manufacturing systems are approached from a bottom-up perspective in an attempt to reduce the specification of the production processes to the minimum.

Modern manufacturing companies are betting on the application of intelligent and more integrated monitoring and control solutions to reduce maintenance problems, production line downtimes and reduction of manufacturing operational costs while guarantying a more efficient management of the resources and an improved quality of products. The shoe industry provides a fertile ground in this direction since traditionally the production and manufacturing of shoes involves a wide variety of materials and a large number of both operations and machines characterized by a huge number of parameters as well. Thereby, the optimization of manufacturing process parameters during production activities is recognized as one of the most important task. As a matter of fact, the selection of the best set of manufacturing process parameters can improve final product quality, cost effectiveness while reducing anomalous situations that potentially may cause a line stopping. The present paper describes the research background that has driven the design and development of the Self-Learning methodology and reference architecture as the foundation for a new generation of monitoring and control solutions. Furthermore, a real application scenario from the shoe industry is also described to demonstrate the applicability of the proposed solution.

JADE is a multiagent-based framework widely used for prototyping agent-oriented mechatronic architectures. JADE's messaging performance is normally rated unsuitable for real time operation in a mechatronic context and there are a few papers in the literature backing up this claim. The author's share this view however the main purpose of this paper is to highlight that distinct message exchange coding patters influence the round trip time (RTT) of the messages quite substantially. The paper presents a comparative performance analysis, based on the RTT, of these coding patterns. The majority of the published tests is focused in the evaluation of the RTT when pairs of communicating agents are added to the platform. The analysis presented in this work is focused on the efficiency of each method as the responder agent receives messages continuously.

The manufacturing processes of today are caught between the growing needs for quality, high process safety, efficiency in manufacturing process, reduced time-to-market and higher productivity. In order to meet these demands, more and more manufacturing companies are betting on the application of intelligent and more integrated monitoring and control solution to reduce maintenance problems, production line downtimes and reduction of manufacturing operational costs while guarantying a more efficient management of the manufacturing resources. In this scenario, the research currently done under the scope of the Self-Learning Production Systems (SLPS) tries to fill these gaps by providing a new and integrated way for developing monitoring and control solutions based on novel technologies and especially on self-adaptive, context awareness and data mining techniques. This paper introduces the research background that has driven the design of the generic SLPS architecture and focuses on the Adapter component responsible for adapting the system behaviour according to the actual operative context. The proposed Adapter architecture together with its core components are introduced as well as the generic adaptation process, or rather, the way the Adapter adapt the system behaviour to cope with the current context. Finally, to demonstrate the applicability of the SLPS methodology into real industrial context as well as the Adapter capabilities to learn and evolve along system lifecycle an application scenario is presented.

This paper discusses the main problematic, misunderstandings and gaps associated with the development of self-organizing multiagent mechatronic systems. The paper reflects the authors' experience in designing and implementing Multiagent Mechatronic Systems. It also partly addresses the work developed under the FP7 IDEAS project (rated an EU FP7 success story) as a clarifying, but not unique, example. In this respect the paper presents a critical overview on how the existing technology is close to support the automation concepts envisioned by emerging production paradigms that rely on Emergence and Self-Organization as main constructs of the system and why technology alone should be understand as only a mean and not a end.

There is a substantial difference between traditional industrial systems and systems resultant from emerging production paradigms in terms of both conceptualization and implementation. Backwards compatibility is fundamental in assessing the potential transitional period whereby legacy technology will operate under the conceptual framework of modern approaches. However, before this point can be reached it is necessary to further investigate, assess and quantify the behavior of modern approaches that increasingly rely in self-organization. This article's goal is to give an overview of the behavioral assessment problematic, in the context of self-organizing mechatronic systems, and provide a discussion on opportunities to explore rule-extraction techniques to better understand the influence of design and configuration on the overall system behavior. Although the discussion widely applies to almost all emerging production paradigms the Evolvable Production System paradigm is used as case to bring context and clarity to the discussion.

With the emergence of new modern manufacturing paradigms new concepts, originally from the complexity sciences started to be introduced in the manufacturing systems, rendering traditional control approaches insufficient. Therefore, new approaches were developed, supported by the modern manufacturing paradigms bio-inspired background. However, somehow along the way the physical and logical nature of the system was partially lost, leading to the convergence of approaches towards more traditional systems, 'neglecting' their bio-inspired principles. With the present work the authors aim to introduce and analyse two new different self-organising approaches that try to bring the focus of manufacturing systems, again to the bio-inspired principles. For this purpose, in the context of this work, manufacturing systems are approached from a bottom-up perspective, in an attempt to reduce the specification of the production processes to the minimum and foster the production emergence. A test case is considered, to draw initial conclusions.

Testing and debugging real hardware is a time consuming task, in particular for the case of aquatic robots, for which it is necessary to transport and deploy the robots on the water. Performing waterborne and airborne field experiments with expensive hardware embedded in not yet fully functional prototypes is a highly risky endeavour. In this sense, physics-based 3D simulators are key for a fast paced and affordable development of such robotic systems. This paper contributes with a modular, open-source, and soon to be freely online available, ROS-based multi-robot simulator specially focused for aerial and water surface vehicles. This simulator is being developed as part of the RIVERWATCH experiment in the ECHORD european FP7 project. This experiment aims at demonstrating a multi-robot system for remote monitoring of riverine environments.

As automated manufacturing systems become more and more complex, the need for new methodologies to improve the design and development of industrial monitoring and control solutions is becoming peremptory. Programmable Logic Controllers (PLCs) dominates the application domain meaning that they are established as the most popular industrial controllers used in factory and shop floor. Although the capabilities of these controllers have strongly improved in the last decades, their historical background as the easy understand by the electricians who had previously worked on electrical systems has meant that the de-facto standard for the implementation of control and monitoring solution for these devices remains the ladder logic. However, to face the globalization challenges, the manufacturing companies needs to improve their productivity by reducing the costs, delivering high-quality of products with high variety and improving their responsiveness to changing market condition. Moreover productivity gains can also be achieved by reducing the life-cycle costs of manufacturing production systems implying the using of more flexible and agile approaches. In this scenario, the proposed methodology aims to provide a homogeneous and optimum process that, starting from a set of mechanical specifications and behavioural models of machines, enable the generation of industrial logic automatically ensuring the same structure and naming standard in every project and the quality of the code.

Due to the growing demand to reduce the environmental impact, the manufacturing companies of today are encouraged to adopt new green methodologies, strategies and technologies for increasing the energy efficiency of their manufacturing production lines. These solutions have a great impact on several productivity metrics including availability and costs. The continuous pursuit of productivity and particularly of machine availability has led to an increase of the total energy consumption in production plants. However, productivity gains can also be achieved by reducing the life-cycle costs of the manufacturing production systems. The research currently done under the scope of Self-Learning Production Systems (SLPS) tries to fill the gap between availability and efficiency by providing an innovative and integrated approach for ensuring the efficient utilization of the resources in machine tools.

This paper details the FP7 IDEAS project multiagent mechatronic architecture. In particular the technical details that support plug and produce of individual modules or stations are discussed. To highlight the generic approach of the core agent-based mechatronic platform an example where a virtual station is introduced and seamlessly interacts with the physical systems is highlighted. This example has, therefore, the dual purpose of showing the generic aspects of the reference architecture, and its implementation, and describing the technicalities that support the introduction of virtual components in the "loop" of the physical systems which goes far beyond mere simulation of assembly systems. This flexibility opens wide the door for improved system analysis where just specific components have to be simulated whereas the remaining dynamic is the one of the real system.

The paper presents the key results of the EU-project Self-Learning. The strategic objective of Self-Learning is to strengthen EU leadership in production technologies in the global marketplace by developing innovative self-learning solutions to enable tight integration of control & so called secondary processes (e.g. maintenance, energy efficiency) of production systems. The project developed highly reliable and secure service-based self-learning solutions aiming at that integration. Approaches based on SOA principles, using distributed networked embedded services in device space, are the most appropriate for implementation of such self-learning solutions. Context awareness, providing information about the processes & equipment and circumstances under which the services operate and allowing them to react accordingly, is a promising holistic approach to assure needed self-learning adaptation to changes in processes and equipment states. The key components of a self-learning solution include a Context Extractor module to allow for dynamic context extraction, processing and storage, an Adapter module to allow for a holistic process control of considered systems and a self-learning module to allow for self-learning. Three industrial application scenarios drive the project.

This paper presents ARES-III, a multi-purpose service robot for robust operation in all-terrain outdoor environments. Currently in pre-production phase, ARES-III is aimed to fulfil the requirements of a robotic platform that is able to support the development of real world applications in surveillance, agriculture, environmental monitoring, and other related domains. These demanding scenarios motivate a design focused on the reliability of the mechanical platform, the scalability of the control system, and the flexibility of its self-diagnosis and error recovery mechanisms. These are key features of ARES-III often disregarded in current commercial and research platforms. First, a comprehensive set of field trials demonstrated the ability of the ARES-III chassis, made of durable materials and with no-slip quasi-omnidirectional kinematic characteristics, to perform robustly in rough terrain. Second, supported by a control system fully compliant with the wide spread Robot Operating System (ROS), the scalability of ARES-III is enforced. Finally, the integration of active self-diagnosis and error recovery mechanisms in ARES-III control system fosters long lasting operation.

In recent years a set of production paradigms has emerged to circumvent pressing production challenges. In particular the ability of composing lower order components to form higher order entities with more complex functionalities and reuse them to build yet more complex system has been widely explored. However material handling aspects have been tackled independently of the control/reconfiguration architecture. Traditionally transport resources have been assumed passive and the organization of the production has been treated as a scheduling problems that often ignores disturbances in the transport system. This renders most of the scheduling approaches too theoretical to be applied in real production scenarios and has motivated researchers to pursuit alternative approaches to this sort of production organization problems. In this paper a technical agent-based architecture that integrates control/reconfiguration aspects with material handling is detailed. The architecture supports the simultaneous usage of conveyor and AGV-based material handling systems and the addition of production or material handling resources is handled in a transparent way through careful generic interaction design. Overall the proposed architecture can be applied to any mechatronic system favouring re-configuration and minimizing reprogramming.

There has been an increasing interest from industry in distributed architectures since they promote a plug-and-produce and robust environment, where adaptability and fault tolerance are native. Much research has been conducted in this field mainly supported by multi-agent and service oriented technologies. Nevertheless the retrieval and visualization of information in distributed systems is a relatively unexplored area. Although the dynamic nature of the multi-agent systems allows gathering information in a prompt manner, doing so might affect the performance of the mechatronic agents. In this sense, the present paper details the architecture of a visualization tool that introduces a reliable but non-invasive approach to retrieve data from distributed platforms as well as a new way to visualize and interpret the information gathered from mechatronic based systems.

A cornerstone of modern production paradigms is the encapsulation of heterogeneity and system complexity. This functional encapsulation is supported by distinct architectural building blocks that once instantiated abstract, manage and control a specific system. Modern production approaches rely in semi hierarchical architectures to organize the interactions between these blocks. In runtime this specific structure promotes a self-organizing response resulting from concurrent interaction between the system's components. This paper simulates the execution of processes in a network of mechatronic agents comprising products, processes' abstractions and resources and analyses the impact of the typical network structure envisioned by modern production paradigms in respect to the time consumed in the production process.

A new approach for the realisation of self-learning production systems based on a context aware approach, allowing self-adaptation of flexible manufacturing processes in production systems, is presented. The usage of dynamically changing context for adaptation of flexible manufacturing lines/processes is proposed. The presented solution includes services for context extraction, adaptation and self-learning allowing high adaptation of production systems depending on the identified context. A generic architecture following Service Oriented Principles is presented allowing for integration of the proposed solution into various production systems. A successful application of the developed solution in real world manufacturing environment is presented.

Contemporary industrial automation struggle to cope with the continuous demand for high quality customized products at low cost at the same time that production lifecycle history proves that unexpected behavior arises, equipment fails and requirements are constantly adjusted. All this denotes a permanent need to reengineer in an fast and transparent manner in order to truly implement an agile factory. The present work includes a reference architecture and device model that tries to exploit Internet-based business models alongside Service-oriented Architecture principles and technology to assist the systems integrator during lifecycle reengineering interventions. All architecture elements are portrayed according to their roles and specificity being mutually interoperable to allow a customized mould according to actual system requirements, constraints and goals. The proposed device model follows above architecture guidelines and it supports the enhancement of the overall system ability in terms of agility and openness by laying over open web standards. It allows the transparent deployment of new services besides built-in services that support the discovery, identification, configuration and monitoring of a particular device.

This paper presents a video-based early fire detection system. Focus is given to the challenges related to the actual deployment of the vision system. Most importantly, background subtraction is performed in a windowed way for improved accuracy, an attentive mechanism is employed to focus a computationally expensive frequency analysis of potential fire regions, interaction with a people detection and tracking system is included so as to enable model-based false alarms rejection, a new colour-based model of fire's appearance as well as a new Wavelet-based model of fire's frequency signature are proposed, and the camera-world mapping is approximated according to a GPS-based learning calibration process. An average success rate of 92.7% at a processing rate of 10 Hz shows the applicability of the model to real-life applications.

Nowadays, Service-oriented Architecture (SOA) paradigm is becoming a broadly deployed standard for business and enterprise integration. It continuously spreads across the diverse layers of the enterprise organization and disparate domains of application envisioning a unified communication solution. In the industrial domain, Evolvable Production System (EPS) paradigm focus on the identification of guidelines and solutions to support the design, operation, maintenance, and evolution of complete industrial infrastructures. Similarly to several other domains, the crescent ubiquity of smart devices is raising important lifecycle concerns such as device setup, control, management, supervision and diagnosis. From initial setup and deployment to system lifecycle monitoring and evolution, each device needs to be taken into account and easily reachable. The present work exploits the association of EPS and SOA paradigms in the pursuit of a common architectural solution to support the different phases of the device lifecycle. The result is a modular, adaptive and open infrastructure forming a complete SOA ecosystem that will make use of the embedded capabilities supported by the proposed device model. The infrastructure components are specified and it is shown how they can interact and be combined to adapt to current system specificity and requirements. Finally, a proof-of-concept prototype deployed in a real industrial production scenario is also detailed and results are presented.

Modern Production Paradigms, supported by state-of-art developments in IT, present major opportunities for the effective implementation of sustainable shop floors. With an increased focus in seamless system reconfiguration most proposals advocate the use of modular and distributed architectures that are dynamic compositions (adaptable and evolvable structures) of cooperating self-contained entities that interact using complex communication patterns with enhanced semantics. These design principles introduce significant diagnostic challenges. The distributed and contained nature of the information renders collective fault events harder to track. Furthermore, current approaches that are adjusted to a more conservative logic-based control, do not provide an adequate response to this problem. This paper proposes an approach to capture the network dynamics of faults in the emergent domain of Evolvable Production Systems (EPS) which implements and extends the before-mentioned characteristics.

This paper presents a saliency-based solution to boost trail detection. The proposed model builds on the empirical observation that trails are usually conspicuous structures in natural environments. This hypothesis is confirmed by the experimental results, where a strong positive correlation between trail location and visual saliency has been observed. These results are due in part to the proposed extensions to a well known visual saliency computational model. This paper goes further and shows that, with a proper analysis of the saliency information alone, the ambiguity regarding both trail's position and approximate skeleton is reduced to three hypotheses in 98% of the tested natural images. This analysis is performed by a set of agents inhabiting the saliency and feature specific intermediate maps. These agents' behaviours exploit implicit, top-down knowledge about the object being sought in an active way. With the proposed model, computationally demanding accurate trail detectors are able to focus their activity in a fraction of the input image, thus promoting robustness and real-time performance.

Stereo-based off-road obstacle detection is a complex and still open problem. The challenges are in great extent related to computational cost and noise level. Previous work has shown that visual saliency and voting mechanisms are extremely effective in tackling these issues. This paper proposes a set of extensions to these mechanisms, to further improve the detector's speed-accuracy trade-off as well as its robustness. The observed enhancements are due in part to the adaptive way saliency is accounted for by the detector during the image scanning procedure. Additionally, detector's positive results are in turn used to boost the saliency map itself, thus reinforcing the analysis of relevant regions of the image. To enable detection in highly roughed terrain, the detector's invariance to the robot's posture is also enhanced. Experimental results show that, with the extended detector, denominated of ESalOD, higher levels of robustness, accuracy and computational efficiency are attained.

This paper proposes a model for trail detection that builds upon the observation that trails are salient structures in the robot's visual field. Due to the complexity of natural environments, the straightforward application of bottom-up visual saliency models is not sufficiently robust to predict the location of trails. As for other detection tasks, robustness can be increased by modulating the saliency computation with top-down knowledge about which pixel-wise visual features (e.g., colour) are the most representative of the object being sought. This paper proposes the use of the object's overall layout instead, as it is a more stable and predictable feature in the case of natural trails. This novel component of top-down knowledge is specified in terms of perception-action rules, which control the behaviour of simple agents performing as a swarm to compute the saliency map of the input image. For the purpose of multi-frame evidence accumulation about the trail location, a motion compensated dynamic neural field is used. Experimental results on a large data-set reveal the ability of the model to produce a success rate of 91% at 20Hz. The model shows to be robust in situations where previous trail detectors would fail, such as when the trail does not emerge from the lower part of the image or when it is considerably interrupted.

Agile manufacturing requires high responsiveness at all levels of a company, but is especially challenging on the shop-floor level. Evolvable Assembly Systems (EAS) are a solution: agentified modules can be seamlessly integrated into existing systems, or removed at any instant. EAS offer a more flexible solution to automation production, but many system design and integration tasks are still done manually. Our goal is to make EAS increasingly self-managing: 1) to easily and quickly produce a new or re-configured assembly system each time a new product order arrives or each time a failure or weakness arises in the current assembly system and 2) to maintain production also under degraded conditions. This article describes an architecture for self-managing evolvable assembly systems. It involves on-the-fly self-assembly of robotic modules, dynamic coordination of tasks and self-adaptation to production conditions, mainly self-healing and self-optimisation. The architecture exploits self-description of modules, monitored modules behaviour and dynamic policies.

Inspection of an environment is often limited to a single video feed, provenient from a camera assembled on a mobile robot. This is known as the keyhole effect. In addition to this, the operator must often perform several tasks at the same time and specially in cluttered environments it can be hard to navigate the robot and at the same time keep the viewpoint of a camera aligned with a specific point of interest. This paper tackles this problem by proposing two mechanisms for assisting in teleoperation by positioning a pan-tilt camera. The goal consists in improving the teleoperatorpsilas situation awareness. Field trials with a pan-tilt-zoom (PTZ) camera mounted on an all-terrain robot validate the proposed mechanisms.

Human judgement is an integral part of the teleoperation process, heavily influenced by the rate in which commands are given to and the telemetry data is received from the robot. In this paper a multi-agent system for robot teleoperation is proposed. Several entities (e.g. robot or human) can easily be added and removed. In addition, this approach fosters a shared notion of reality to every entity present in the network, through a mechanism similar to the distributed blackboard architecture. This method provided an exchange of information regarding the teleoperation scenario, i.e. perceptual clues. This promoted a study on how perceptual clues influenced the operators’ judgement and performance. The experimental results in a physical suggest that the system is able to guarantee a close interaction between users and robots.

The unstructured nature of cross-country environments poses a great deal of challenges to obstacle detection algorithms. This paper presents an algorithm able to perform obstacle detection with small computational cycles, and extremely robust against outliers. To reduce the computational cost, foveation, i.e. space-variant image resolution adaptation, is employed. To reduce sensitivity to outliers, two voting filters are presented. Experimental results on two different stereo vision heads show that the proposed algorithm is able to perform at 10 Hz. Even in the presence of a low signal-to-noise ratio, caused by the use of a partially damaged lens, the algorithm is able to distinguish obstacles from the background.

This paper describes the DSAAR architecture, which aims the fast development and prototyping of multi-robot systems and it is based on Linux inter-process communication (IPC) mechanisms and multi-agent Jade software platform. This architecture is of special interest in domains where several heterogeneous actors have to interact both with and without realtime constraints. One of these domains is humanitarian demining, where technology transfer is also a requirement.

This article covers research work about unmanned helicopters that is being developed by the Portuguese SME Introsys, S.A. in the context of a generic project targeting the development of a multi-robot system for landmine detection. As helicopters are complex multivariable nonlinear underactuated systems with strong coupling in some control loops, they are an interesting subject of research for the control and robotic research communities. Therefore, a wide range of techniques have been proposed in the literature to control such vehicles, ranging from classic control to soft-computing methods. The main goal for this article is to present a survey about existing low level control architectures for unmanned helicopters and to discuss which control design criteria (design requirements) are relevant when they are operating in the context of humanitarian demining

This paper describes the development of a multi-robot system for area reduction in humanitarian demining. In spite of the specific requirements imposed to the work being carried out by the humanitarian demining domain, the results being achieved can also be used in other domains such as surveillance and remote monitoring. A multi-agent based architecture is responsible for coordinating a progressive stochastic analysis of the terrain. The all-terrain navigation used by the mobile robots is attained by a novel embodied reactive obstacle avoidance method. A control mission software has been developed to plan, configure and supervise the operations. Highly compliant legged and wheeled platforms have been developed accomplishing low-cost all-terrain robots. Digital signal processing algorithms have been applied for landmine detection using the payload sensors.

Control systems development for the automotive industry is done pretty much the same today as it was 15 years ago. Therefore, the usefulness of a high level structured approach for Control Systems Engineering is undeniable. This paper describes how object-oriented programming (OOP) or the frame paradigm can be used to program PLCs even though using their old fashioned programming abstract model. To prove the validity of the proposal a real industrial implementation is shown.

A generic architecture for evolutive supervision of robotized assembly tasks, in a context of integrated manufacturing systems, is presented. This architecture provides, at different levels of abstraction, functions for dispatching actions, monitoring their execution, and diagnosing and recovering from failures. The problem of integration of legacy systems is discussed and an implementation approach described. Modeling execution failures through taxonomies and causal relations plays a central role in diagnosis and recovery. Through the use of machine learning techniques, the supervision architecture will be given capabilities for improving its performance over time. Particular attention is given to the inductive generation of structured classification knowledge for diagnosis. Methodologies used, performed experiments, and obtained results are described in detail.

A generic architecture for execution supervision of robotic assembly tasks is presented. This architecture provides, at different levels of abstraction, functions for dispatching actions, monitoring their execution, and diagnosing and recovering from failures. Modeling execution failures through taxonomies and causal networks plays a central role in diagnosis and recovery. A discussion on the process of acquisition of such monitoring knowledge is made. Through the use of machine learning techniques, the supervision architecture will be given capabilities for improving its performance over time. Preliminary results of applying machine learning in this area are presented and planned extensions discussed.>