SuPerWorld:Pervasive Real-World Computing for Sustainability

Research / Research Groups / SuPerWorld:Pervasive Real-World Computing for Sustainability
The “Pervasive Real-World Computing for Sustainability” (SuPerWorld) Multidisciplinary Research Group (MRG) involves technology in pervasive computing, which touches upon research that focuses on the real world towards (contributing in) making it more sustainable. As “real world”, we refer to humans, animals and plants, as well as to the physical and urban environment. As “pervasive computing”, we refer to sensing infrastructures and equipment that measure the real world, including the sensing capabilities of modern mobile phones. An important goal would be to enable pervasive eco-systems by means of “Internet of Things” (IoT) and “Web of Things” (WoT) protocols and principles, towards common understanding and semantics of real-world services and data, and high interoperability among systems and infrastructures.

The main research question of the SuPerWorld MRG can be summarized as: “how can pervasive computing affect the real world, in order to promote sustainability, enhance quality of living for humans and animals, ensure food safety and security, as well as provide convenience to humans in their everyday life"? The general domains of the SuPerWorld MRG are “smart cities” and “sustainable environments”. More specific application areas involve food safety, smart agriculture, disaster monitoring, environmental and wildlife monitoring, legal migration, tourism and promotion of digital heritage, urban landscapes and infrastructures, robotics, as well as the future smart grid and energy communities.

Some emphasis is given to the environment of Cyprus, and the particular challenges it faces (e.g. endangered animal and plant species, deforestation, water quality of the coastal areas). Due to the particular sociopolitical situation of Cyprus, some priority is given also to vulnerable target groups (e.g. immigrants and refugees). The SuPerWorld MRG performs basic and (mostly) applied research in the scientific areas of machine learning and deep learning, geospatial analysis and Geographical Information Systems (GIS), IoT and WoT, data visualization, semantic web technologies, remote sensing (aerial photography based on drones and satellite-based imagery), mobile phone computing, robotics, as well as persuasive technology, ICT for sustainability and eco-feedback techniques for human behavioural engineering. Finally, the SuPerWorld MRG supports, encourages and promotes a lively ecosystem of interdisciplinary collaboration, which constitutes one of the strong aspects of RISE.
MRG leader:
Dr. Andreas Kamilaris

Savvas Karatsiolis
Expertise: Computer vision, Deep learning, Machine learning

Lanfa Liu
Expertise: Remote sensing, Satellite imagery
Ian Cole
Expertise: Renewable energy, GIS, Solar and photovoltaic systems

Olivia Guest
Expertise: Human cognition modelling, Deep learning

Our team has a multi-disciplinary expertise, know-how and experience, covering a wide spectrum of sensory-based computing and modelling. Our group has a specific research interest in large-scale earth observations via remote sensing (resp: Lanfa), using Internet of Things field sensors as ground truth data (resp: Andreas), employing computer vision via deep learning for classification and prediction (resp: Savvas),  as well as geospatial analysis afterwards (resp: Andreas) for modelling and problem solving, towards more informed policy-making. An application of this analysis towards policy-making is in the field of renewable energy systems (resp: Ian). Finally, our team tries to dig deeper into deep learning models, trying to improve their operation based on how human cognition works (resp: Savvas, together with Olivia). Our team structure based on know-how is depicted below.


Application domains:
Agriculture, Food Supply Systems, Smart Cities and Urban Environments, Robotics, Forestry, Ecology, Renewable Energy and Smart Electricity Grid.




Identifying invasive species in Janapese mixed forests

The goal is to use aerial photos coming from drones and/or satellite images, combining them together with ground truth data, in order to teach deep learning models to correctly classify tree species in Japanese forests based on their canopy. The long-term aims are to map and quantify Japanese mixed forests, identifying invasive species and their impact to the overall flora of the areas under study.

Advanced deep learning models are being used (i.e. RNN, LSTM), capturing satellite images and aerial photos from drones in different seasons of the years, allowing to address the problem with high precision.
Collaboration with: Prof. Larry Lopez, Sarah Kentsch (Yamagata University, Japan)
Techniques used: Satellite imagery, aerial imagery (UAVs, drones), deep learning, Internet of Things



illegal_dumping.jpgIdentifying illegal dumping from satellite imagery

Illegal dumping is a frequent phenomenon in south European countries, including Cyprus. Dumping involves organic material, construction waste, old electronichs and machinery, abandoned cars etc. The goal of this project is to identify where dumping occurs from satellite imagery and then check whether this dumping is illegal.
Spots of dumping will be indicated by citizens through their mobile phones, via a crowdsourcing-based approach.
The identification system will then be automatically connected to the respective municipality services, depending on the exact location where dumping has occurred, in order to take measures. The case study takes place in Cyprus, with a strong focus in the sub-urban areas of Nicosia.
Collaboration with: CLEANathon team (Cyprus), Friends of the Earth (Cyprus)
Techniques used: Satellite imagery, deep learning, mobile phone computing (citizens’ crowdsourcing), geospatial analysis, GIS




Animal manure as fertilizer for crop fields

Intensive livestock production might have a negative environmental impact, by producing large amounts of animal dejections, which, if not properly managed, can contaminate nearby water bodies with nutrient excess. However, if the animal manure could be transferred to nearby crop farms, to be used as a fertilizer for the crops, then the problem of pollution/contamination would be mitigated, transforming manure from a waste to a resource. This valorization of manure from waste to a resource falls within the principles of circular economy, but the transportation of manure also comes at an environmental and economic cost. It is a single-objective optimization problem, in regards to finding the best solution for the logistics process of satisfying nutrient crops needs by means of livestock manure.
This project investigates a centralized optimal algorithm to solve the problem, based on a realistic simulator that considers numerous real-world constraints, which have not been addressed by related work. Case study is performed in Catalonia, Spain, in agreement with the Department of Agriculture, Government of Catalonia.
Collaboration with: Francesc Prenafeta (IRTA Barcelona, Spain)
Techniques used: Geospatial analysis, GIS, Internet of Things, big data analysis, modelling, simulations.


Blockchain in agriculture and food supply systems

blockchain_agri.jpgBlockchain is an emerging digital technology allowing ubiquitous financial transactions among distributed untrusted parties, without the need of intermediaries such as banks.
The aim of this project is to examine the impact of blockchain technology in agriculture and food supply chain, studying existing ongoing projects and initiatives, understanding overall implications, challenges and potential, with a critical view over the maturity of these projects.
Our current findings indicate that blockchain is a promising technology towards a transparent supply chain of food, with many ongoing initiatives in various food products and food-related issues, but many barriers and challenges still exist, which hinder its wider popularity among farmers and systems. These challenges involve technical aspects, education, policies and regulatory frameworks.
Collaboration with: Francesc Prenafeta (IRTA Barcelona, Spain)
Techniques used: Bibliographic analysis, survey.
Publications: Kamilaris, A., Fonts, A. and Prenafeta-Boldύ, F.X., 2019. The rise of blockchain technology in agriculture and food supply chains. Trends in Food Science & Technology, 91, pp.640-652.
 [ ]



Initiated by the Paris Agreement of December 2015 is the 2050 low-carbon economy roadmap. Relative to 1990 levels, the greenhouse gas emission reduction milestones associated with this roadmap are: 40% by 2030, 60% by 2040 and 80% by 2050. The power sector has the biggest potential for emissions reductions, not only does the sector have the potential to almost eliminate CO2e emissions by 2050 but the electrification of other sectors, such as transportation, can further reduce emissions in general.
To address these issues, the power sector has been undergoing significant change in recent years. Renewable energy technologies have seen wide scale adoption, most significantly of wind and solar photovoltaics installation at the utility scale, decarbonizing the energy mix yet presenting integration problems due to their associated variability, unpredictability and asynchronicity of supply. Small-scale renewables have seen less adoption and are typically a lot less reliable. The adoption of small-scale renewables tends to include relatively high up-front costs for viability considerations and the relative cost of monitoring equipment is generally economically prohibitive. However, small-scale localized photovoltaic installations empower the general population by increasing their self-sufficiency and energy security.
The future electricity network is a network of distributed prosumers, self-consuming their generation when possible and drawing from or feeding into the network when not. Adoption of small-scale renewable energy generation systems also leads to increased human capital in renewable energy, raising awareness and understanding of the energy picture in general. A well-developed and freely available tool for automated analysis can accelerate the deployment of small-scale photovoltaic systems whilst informing the distribution and transmission system network operators of the effects, allowing them to implement mitigation strategies.
Collaboration with: Ian Cole (RISE, Cyprus)
Techniques used: Satellite imagery, geospatial analysis, GIS, Internet of Things, big data analysis, modelling, simulations.
Publications: Palmer, D., Koubli, E., Cole, I., Betts, T. and Gottschalg, R., 2017. Comparison of solar radiation and PV generation variability: system dispersion in the UK. IET Renewable Power Generation, 11(5), pp.550-557. ]




 Internet of Things in robotics

Untitled.pngAs the Internet of Things (IoT) penetrates different domains and application areas, it has recently entered also the world of robotics. Robotics constitutes a modern and fast-evolving technology, increasingly being used in industrial, commercial and domestic settings. IoT, together with the Web of Things (WoT) could provide many benefits to robotic systems. Some of the benefits of IoT in robotics have been discussed already in the past. This project moves one step further, studying the actual current use of IoT in robotics, through various real-world examples encountered through a bibliographic research. The project also examines the potential of WoT, together with robotic systems, investigating which concepts, characteristics, architectures, hardware, software and communication methods of IoT are used in existing robotic systems, which sensors and actions are incorporated in IoT-based robots, as well as in which application areas.
Finally, the current application of WoT in robotics is examined and discussed.
Collaboration with: Nicolo Botteghi (Robotics and Mechatronics Group, University of Twente, Netherlands)
Techniques used: Bibliographic analysis, survey, Internet of Things, robotics/mechatronics.
Kamilaris, A. and Botteghi, N., 2020. The Penetration of Internet of Things in Robotics: Towards a Web of Robotic Things. arXiv preprint arXiv:2001.05514.
 [ ]


Swarms of robots in agricultue

Robotics is expected to play a major role in agriculture, and multi-robot systems and collaborative approaches can constitute potential solutions to improve efficiency of agricultural operations and processes. This field of research is called swarm robotics, and it is about coordinating multiple robots as a unified system, which consists of large numbers of individual physical robots. It is supposed that a desired collective behavior emerges from the interactions between the robots, as well as from the interactions of robots with the environment.
Swarms of robots tackle challenges of flexibility, scalability and robustness in solving complex tasks related to precision farming. An important problem of robotic swarms is how to dynamically divide some agricultural field and split the operations needed to be performed, which can be weed detection, seeding, estimation of crop yield and others. The goal here is to find a dynamic solution for the division of labour among robots, independently of the structure of the agricultural field. Relevant performance metrics include the overall time to finish the job, the least energy consumed by the robots (i.e. avoiding overlaps in trajectories) and the coverage of the whole field (i.e. coverage ratio). Robots can be either ground robots or unmanned aerial vehicles (UAV)/drones.
Collaboration with: Nicolo Botteghi (Robotics and Mechatronics Group, University of Twente, Netherlands), Beril Sirmacek (Jönköping University, Sweden)
Techniques used: Geospatial analysis, GIS, Internet of Things, big data analysis, modelling, simulations.




How can insights from biological brains and behaviour be applied to machine learning?

Cognitive science and artificial intelligence have an intertwined history. For example, more than half a century ago neural networks emerged from the interdisciplinary work done by psychologists and computer scientists — David Marr’s deep insights into vision and levels of analysis are a function of his knowledge in both of these broad scientific domains. In this project, we will use insights from human and animal cognition — i.e., that representations found in the brain have certain computational properties, and that the categorisation literature in children and adults provides certain frameworks for successful learning and generalisation. We will translate and apply these findings with the aim of improving deep learning models in terms of transparency, accountability, and training.
One part of this project involves using the alignment of qualitatively different sources of data (e.g., images, words, and audio) to create richer multi-modal semantic spaces in deep neural networks. What this means is that we can take one type of data, like images, and compare and combine the space the model makes for these with the representational spaces that already exist for, e.g., words. This will pave the road towards better machine learning models as well as better models of human and animal cognition.
Collaboration with: Olivia Guest (RISE, Cyprus)
Techniques used: Deep learning, modelling.



Subscribe to our Newsletter

* indicates required