Ebook: Intelligent Environments 2020

Intelligent Environments (IEs) aim at empowering users by enriching their experience, raising their awareness, and enhancing their management of such environments. IEs refer to the physical spaces where ICT and pervasive technologies are used to achieve specific objectives for the user and/or the environment. IEs’ growing community, from academia to practitioners, is working on the materialization of IEs driven by the latest technological developments and innovative ideas.

The 16th International Conference on Intelligent Environments focuses on the development of advanced intelligent environments, as well as newly emerging and rapidly evolving topics. The present edition has been impacted by the COVID-19 pandemic, and has forced us all to celebrate it virtually. We are pleased to include the proceedings of the following workshops, which emphasize multi-disciplinary and transversal aspects of IEs, as well as cutting-edge topics:

• 10th International Workshop on Intelligent Environments Supporting Healthcare and Well-being (WISHWell’20)

• 9th International Workshop on the Reliability of Intelligent Environments (WoRIE 2020)

• 4th International Workshop on Legal Issues in Intelligent Environments (LIIE’20)

• 4th International Workshop on Intelligent Systems for Agriculture Production and Environment Protection (ISAPEP’20)

• 4th International Workshop on Citizen-Centric Smart Cities Services (CCSCS’20)

• 2nd International Workshop on Intelligent Environments and Buildings (IEB’20)

• 1st International Workshop on Research on Smart Grids and Related Applications (SGRA’20)

• 1st International Workshop on Open and Crowdsourced Location Data (ISOCLoD’20)

• 1st International Workshop on Social Media Analysis for Intelligent Environment (SMAIE’20)

The proceedings contain a series of contributions reflecting the latest research developments in IEs and related areas, focusing on stretching the borders of the current state of the art and contributing to an ever-increasing establishment of IEs in the real world. We would like to thank all the contributing authors, as well as the members of the Organizing Committees and Program Committees of the workshops for their highly valuable work, which contributed to the success of the Intelligent Environments 2020 Event.

We are grateful to our technical sponsors, the IEEE Systems Man & Cybernetics Society, IOS Press and the MDPI journals: Sensors, Electronics and Applied Sciences.

July 2020

The Editors

“In the midst of every crisis, lies great opportunity”, this unverified quote, often attributed to Albert Einstein, is a hopeful statement in those early months of 2020. While, the dramatic human and economic costs of the COVID-19 crisis are still unraveling, there is at the same time the immense human push to create the path out of this crisis, to challenge our habitual patterns, and to envision solutions for the future. This drive towards innovation is very apparent in the area of intelligent environments, where we see an increased movement towards solution that support the monitoring of individual and public health, that massively expand the scope of data collection for societal benefits, ultimately trying to rebuild trust between people and nations, as well as accelerating the way to a new normal. In this work I will provide an overview of the most recent developments of innovative solutions in intelligent environment technologies and how those support the current efforts to overcome the global COVID-19 health crisis.

Currently, society has more aging filters and this is done because: each year the amount of adults who exceed 60 years of age increases, each year more men and women participate in activities outside their homes and thanks to this, new solutions are required to help take care of older people (the elderly). Systems that support us with the care of the elderly at home already exist and some are economic, allowing healthcare centers to obtain them and help them remotely take care of our love ones. As the years go by, technology has made the comfort of our loved ones at home possible. In this research, a novel architecture and initial prototype is presented focused on comfort and it exploits the multi-agent systems paradigm that includes both stationary and mobile agents. In addition, an adaptative mechanism is presented that allows mobile agents to adapt to diversified local environments.

An expanding attention regarding human emotion is a pressing motive towards the current research in neuroscience and artificial intelligent. People need to communicate by exchanging information through verbal or nonverbal communication via sound, visual gestures (facial expression or hand/body gestures). In today’s society, digital multimedia is one of the essential elements in daily life activities that can emphasize communication and emotions adequately. People with severe motor disabilities have difficulties in communicating and showing their emotions directly. Therefore, brain computer interface (BCI) can be a helpful tool as an alternative and assistive communication tools for sharing emotional information. This paper has conducted a review analysis to present the current trend in using digital multimedia to express the human feelings for the latest five years. Twenty-nine studies were selected from IEEEXPlore, PubMed and ScienceDirect, and classified into three major categories: methodology, multimedia type and number of emotion classes. The results show the need for more case studies and games in this area. There is also a need to increase the quality and quantity of research in emotion using the electroencephalography (EEG).

Remote rehabilitation systems allow the supervision and monitoring of physical exercises by therapists without the need to move, temporally and spatially, the patients who perform them. The main advantage of this approach is the patient’s increased autonomy and flexibility to carry out rehabilitation from home, especially in situations of lock-down and movement restrictions. In order to make the execution of repetitive exercises more dynamic and to motivate patients to perform them from home, in recent years gamification techniques, exergames, and serious games have been extensively used. In this context, and to increase the remote monitoring capabilities of therapists, this paper proposes the use of a language for the specification of exergames oriented to the definition, by therapists, of key performance indicators and mobility constraints adapted to the rehabilitation process of each patient. The sentences of this language can be processed by software, allowing the automatic generation of personalised games for rehabilitation. A case study describing an exergame for the upper limb rehabilitation of stroke patients is also presented.

This paper presents a system consisting of a smart medicine dispenser of solid medications (pills, capsules, …) and a mobile application for its configuration and management. The main idea is to offer a solution to help people (especially vulnerable ones) to avoid incorrect medication intakes. In this regard, the smart dispenser delivers the required medication if two conditions are met: (1) it is the scheduled time for a medication intake, and (2) the person who removes the medication from the dispenser (patient or caregiver) can be identified and is authorized to do so. Person identification and authorization is performed through facial recognition by the dispenser and through a username and a password by the mobile application. Moreover, the system reminds the users whenever a medication intake should take place through mobile notifications and lights and sounds emitted by the dispenser. The system development has been guided by a Test-Driven Development Methodology for Internet of Things (IoT)-based Systems to promote its quality and reliability.

Nowadays we can witness many mobile and IoT devices constituting intelligent environments. The physical status of the real world is obtained with many wirelessly-connected devices and it seems that we can create any arbitrary intelligent environments. However, these devices still have a problem of inflexibility of supplied power; wireless IoT devices tend to necessitate the batteries solely dedicated to them. Therefore, as the number of IoT devices increases, the cost of managing these batteries also increases. To cope with the problem, Ichikawa et al. has developed Virtual Grid Hub (VGHub), where USB-PD-connected devices can alternatively provide and receive electrical power. In this study, we show how to control VGHub connecting personal computers, mobile phones, and portable batteries to maximize the utilization of electrical power taking practical issues into consideration.

With the growth of the use of embedded systems in safety-critical applications, the demand for predictable and reliable real-time systems has increased drastically. A large percentage of real-time systems developed today are still built using C due to the performance requirements, and hence inherently unsafe. The advent of Rust has made it possible to achieve safety and reliability without any compromise on performance. This paper presents HarSaRK-RS, a priority-based preemptive hard real-time kernel implemented in Rust. The proposed kernel design and architecture ensure safety at compile time keeping the data-structure and runtime overhead of the kernel minimal, thus enhancing the real-time guarantees of the system. It guarantees freedom from data races, deadlocks, and priority inversion at compile-time. The Kernel core is independent of any clock for its operation, making it power efficient and ideal for battery-operated environments.

Deep learning combined with autonomous drones is increasingly seen as an enabler of automated aircraft inspection which can support engineers detect and classify a wide range of defects. This can help increase the accuracy of damage detection, reduce aircraft downtime, and help prevent inspection accidents. However, a key challenge in neural networks is that their stability is not yet well understood mainly due to the large number of dimensions and the complexity of their shapes. This paper illustrates this challenge through a use case that applies MASK R-CNN to detect aircraft dents. The results show that environmental factors such as raindrops can lead to false positives. The paper also proposes various test scenarios that need to be considered by the developers of the drone-based inspection concept to increase its reliability.

In general, software agents deployed in a multi-agent systems (MASs) collaborate and share data with each other by means of Foundation for Intelligent Physical Agents (FIPA) communications into the same agent platform (intra-). Java Agent Development framework (JADE) provides a facility software framework to build software agents executing in a specific agent platform. This paper presents a novel agent communication gateway aimed at abstracting FIPA communication between software agents on external (inter-) platforms. The novel agent communication gateway is based on the Agent-as-a-Service model (AaaS) implemented using REST technologies. The use of this gateway can favor and enable the interoperability among agents belonging to different MASs, opening collaborations and interactions among heterogeneous agents everywhere in Internet, whenever FIPA message exchanges are used as standard. In this paper some communication paradigms are presented based on the proposed gateway. Some implementation issues show that the programming complexity of communication blocks of agents can be decreased at both Java applications and Java web systems.

IoT for the house raises several concerns under the EU Competition and IP law. In order to create a better level of interoperability between the different smart home IoT objects, undertakings are willing to create new standards, in particular concerning the Internet Protocol. However, this problem seems to be over. In December 2019, Google, Amazon, Apple, Ikea, the ZigBee alliance and many other undertakings belonging to the electronics and innovation world joined together to form the project “connected home over IP” whose main goal is to create “…a new, royalty free, connectivity standard”. This paper aims at giving a first analysis of this project under a competition law and IP law perspective. The main focus will be both on the interplay between Art.101(1),(3) TFEU and on Standard Essential Patents (SEPS) applied to this project. It will be argued that if this project is considered a collusive agreement, it could become a new case in the SEP litigation framework.

The exponential spreading and deployment of emerging digital technologies such as the Internet of Things (IoT) has been remarkable: the IoT market is expected to triple, at least, from USD 170.57 billion in 2017 to USD 561.04 billion by 2022. IoT technologies collect, generate and communicate a huge amount of different data and metadata, through an increasing number of interconnected devices and sensors. Current EU legislation on data protection classifies data into personal and non-personal. The paper aims at charting the resulting entanglements from an interdisciplinary perspective. The legal analysis, integrated with a technical perspective, will address firstly the content of IoT communications, i.e. “data”, and the underlying distinction between personal and non-personal. Secondly, the focus will shift on the metadata related to communications. Through a technical analysis of the highly sensitive nature of metadata, even when the content is encrypted, I will argue that metadata are likely to undermine even more the ontological and sharp division between personal and non-personal data upon which the European legal frameworks for privacy and data protection have been built. The incoming ePrivacy Regulation shall provide metadata, which should be considered always personal data, the same level of protection of “content” data. This interpretation might broaden the scope of application of GDPR and the connected obligations and responsibilities of data controllers and data processors too much.

The paper focuses on technological and legal impact of the Covid-19 crisis, drawing the attention to the convergent nature of Big Data in healthcare emergencies. While, Big Data tools have the potential to prevent, predict, manage or treat arising pandemics, at the same time the use of data driven solutions as raise numerous privacy concerns. European General Data Protection Regulation (GDPR) may fall short when it comes to addressing risks of Big Data analytic tools, in particular when it comes to the collective dimension of data protection rights. However, as shown by the recently issued guidelines of the European Data Protection Board (EDPB) is possible to find a “fair-balance” between public authorities and private entities necessity to implement surveillance measures and the respect of data protection fundamental principles.

Tomato is one of the most grown and the second most consumed vegetable in the world. Alternaria solani is recognized as the most dangerous tomato pathogen. Currently, the diagnostics of this disease requires the proper symptoms assessment of plant tissue damages. Therefore, it is necessary to propose a fast and reliable method able to assess the degree of plants’ damage. Hyperspectral measurements and machine learning algorithms are one of the possibilities to address the problem of finding fast and even more important nondestructive plant diseases detection method. The presented work describes the application of two ensemble learning algorithms: Decision tree and Random Forest adapted for Alternaria solani detection for two varieties of tomatoes cultivated under foil tunnels. The final model was trained on the hyperspectral measurements from 350-2500nm spectral range. With a resulting accuracy of the method:0.78 and 0.98 for decision tree and random forest algorithms, respectively.

The knowledge of spatial variability in soil organic carbon (SOC) is an important consideration in precision agriculture as well as site specific nutrient management. Geostatistical analyses coupled with GIS and GPS are effective tools in assessing the spatial variability and mapping of SOC. A total of 268 soil samples were collected in a systematic grid design (1-minute interval) using GPS covering four sub-districts: Delduar, Melandah, Mirpur and Fultala under two major alluviums - the Ganges and the Brahmaputra. The classical statistics showed that SOC values are normally distributed in the Fultala sub-site whereas in the other sub-sites, the SOC contents were not normally distributed. The semivariogram model also shows that the Fultala sub-site appears to have a strong structure and a gradual approach to the Gausian model providing the best fit where as the other sites show a weak spatial dependency. Due to salinity and other constrains, Fultala sub-site bears a relatively low cropping intensity and hence tillage and crop management are much lower than the other sites. GIS based interpolated values of SOC ranged from 0.39 to 2.02 % in the Fultala sub-site. Interpolated values of SOC ranged from 0.40 to 2.60% in the Delduar sub-site, 0.40 to 1.35% in the Melandah sub-site and 0.38 to 1.39% in the Mirpur sub-site respectively. Clearly, the sites where SOC is low, a pragmatic and location-based policy should be adopted to maximize SOC sequestration. Therefore, the geospatial technologies can help better management of agricultural land by targeting management practices appropriate to the SOC levels.

Nowadays, precision agriculture has acquired an important role in improving the existing problems in this domain, allowing to increase agricultural production and, therefore, the economic profits derived from it. The general objective of precision agriculture is to help and/or improve decision making regarding the quality, productivity, profitability, resource efficiency and sustainability of agricultural production. One way to carry out this objective is to analyze how suitable a type of crop is in the geographical area in which it is located. In this work, a Intelligent Data Analysis process is used and the study that should lead us to obtain knowledge of geographic regions with similar characteristics from climate viewpoint is started. This should allow us to advise farmers in areas with weather conditions that will cause a lower yields, which types of crops will lead to higher yields and could be well suited to their areas because they are getting good yield in zones with similar weather conditions.

Honeybees, in their role as pollinators, are vital to both agriculture and the wider ecosystem. However, they have experienced a serious decline across much of the world over recent years. Monitoring their well-being, and taking appropriate action if that is in jeopardy, has thus become a matter of great importance. In this paper, we present an approach based on computer vision to monitor bee activity and motion in the vicinity of an entrance/exit to a hive, including identifying and counting the number of bees approaching or leaving the hive in a given image frame or sequence of image frames.