Researchers from the University of Southampton are aiming to develop a ‘smart’ thermostat to help UK households save money on their energy bills.

The prototype autonomous device allows users to control their heating based on the price they want to pay rather than setting it by temperature alone, which existing smart thermostats such as Nest, do.

To understand how users interact with smart energy systems, a team of researchers from Electronics and Computer Science (ECS) at Southampton and the University of Zurich produced three different smart thermostats that automated heating based on users’ heating preferences and real-time price variations: a manual one through which participants explicitly specify how the heating should respond to price changes, and two learning-based ones that employed an artificial intelligence (AI) algorithm to automate the temperature settings based on learned households’ preferences.

The designs were tested through a month-long field study involving 30 UK households. The energy price was changed every 30 minutes, which could be viewed by the participants.

The results showed that participants used all three thermostat systems to effectively manage their home heating and create temperature preferences based on real-time prices.

University of Southampton PhD student Alper Alan, co-author of the study, says: “People were more aware of their energy consumption and were happy with the autonomous system controlling their heating on their behalf given real-time prices. We saw that people liked being able to control their home from anywhere they are. We also observed in all three treatments, most people reduced their temperature at high price levels. We found that people are ready to incur a reduction of four degrees celsius to save money. They said that even though the house was a bit colder than normal, they were okay with it since it was not that cold and they were saving money.”

However, following interviews with study participants it was clear that the price was not only the factor that users considered for heating their home. Other key factors were outside weather, occupancy and daily activities within the house. Some participants explicitly stated that the use of the thermostat could be more convenient if it could learn their occupancy patterns. Also, outside weather and the activities that they perform during a day within the house have a significant impact on how people feel the indoor temperature.

Dr Enrico Costanza, co-author from ECS, says: “Future design of learning thermostats should not only take into account occupancy patterns and outdoor temperatures (such as the Nest system), in addition to price preferences, but also people’s daily routines and activities. For instance, we saw that most participants preferred to have the heating on when they shower and have the heating off when they use their oven or perform physical activities.”

The study will be presented at ACM CHI 2016, the top conference for Human-Computer Interaction in San Jose, United States, in May.

Dr Sarvapali Ramchurn, co-author from ECS, adds: “Smart energy systems that use machine learning techniques are increasingly integrated in all aspects of our lives. These findings will assist designers in improving user experience, which in return will help us in controlling home heating with real-time pricing and benefit from them.”

The research was funded by the Engineering and Physical Sciences Research Council (EPSRC) through the ORCHID and CharIoT projects.

Evolution may be more intelligent than we thought, according to a professor in Electronics and Computer Science at the University of Southampton.

Professor Richard Watson of the Agents, Interaction and Complexity research group says new research shows that evolution is able to learn from previous experience, which could provide a better explanation of how evolution by natural selection produces such apparently intelligent designs.

By unifying the theory of evolution (which shows how random variation and selection is sufficient to provide incremental adaptation) with learning theories (which show how incremental adaptation is sufficient for a system to exhibit intelligent behaviour), this research shows that it is possible for evolution to exhibit some of the same intelligent behaviours as learning systems (including neural networks).

A cover story article in New Scientist, looks at research from Professors Watson and Eörs Szathmáry, from the Parmenides Foundation in Munich, explaining how formal analogies can be used to transfer specific models and results between the two theories to solve several important evolutionary puzzles.

Professor Watson says: “Darwin’s theory of evolution describes the driving process, but learning theory is not just a different way of describing what Darwin already told us. It expands what we think evolution is capable of. It shows that natural selection is sufficient to produce significant features of intelligent problem-solving.”

For example, a key feature of intelligence is an ability to anticipate behaviours that that will lead to future benefits. Conventionally, evolution, being dependent on random variation, has been considered ‘blind’ or at least ‘myopic’ – unable to exhibit such anticipation. But showing that evolving systems can learn from past experience means that evolution has the potential to anticipate what is needed to adapt to future environments in the same way that learning systems do.

“When we look at the amazing, apparently intelligent designs that evolution produces, it takes some imagination to understand how random variation and selection produced them. Sure, given suitable variation and suitable selection (and we also need suitable inheritance) then we’re fine. But can natural selection explain the suitability of its own processes? That self-referential notion is troubling to conventional evolutionary theory – but easy in learning theory.

“Learning theory enables us to formalise how evolution changes its own processes over evolutionary time. For example, by evolving the organisation of development that controls variation, the organisation of ecological interactions that control selection or the structure of reproductive relationships that control inheritance – natural selection can change its own ability to evolve.

“If evolution can learn from experience, and thus improve its own ability to evolve over time, this can demystify the awesomeness of the designs that evolution produces. Natural selection can accumulate knowledge that enables it to evolve smarter. That’s exciting because it explains why biological design appears to be so intelligent.”

A University of Southampton academic will participate in the world’s largest project to expand and update our understanding of evolutionary biology.

Dr Richard Watson, associate professor of Electronics and Computer Science and the Institute for Life Sciences at Southampton, is part of an international multi-disciplinary team of 50 world-renowned experts, from eight institutions in the United States, Great Britain and Sweden. The £7.7m project is supported by a £5.7m grant from the John Templeton Foundation, an organisation promoting the advancement of science and philosophy, with a further £2m contributed by the participating institutions.

The collaboration seeks to expand the theory of evolution with new perspectives on the relationships between genes, organism, and environment. It centres on the ‘extended evolutionary synthesis’ (EES) – a new way to think about evolutionary biology aimed at tackling some of its toughest problems. The EES does not replace traditional thinking, but deployed alongside it, aims to stimulate new research within evolutionary biology.

Project leader Professor Kevin Laland at the University of St Andrews said: “The main difference from traditional perspectives is that the extended evolutionary synthesis includes a greater set of causes of evolution. This shifts the burden of explanation for adaptation and diversification; away from a one-sided focus on natural selection and towards the constructive processes of development.”

For example, in the EES, a number of complex biological phenomena are recognised not merely as products of evolution, but as playing a key role in shaping the direction and rate of evolution. For example, in evolutionary developmental biology (evo-devo), the evolution of developmental organisation changes the variation that selection can act on; and in evolutionary ecology (evo-eco), the evolution of ecological organisation changes the selective pressures that act on that variation.

Dr Watson will lead two sub-projects that aim to expand our understanding of both these evolutionary feedbacks using theoretical tools from computer science. “When the products of evolution modify the processes of evolution in this way, this causes problems for current evolutionary theory,” he explains. “But in computer science, these feedbacks are well-understood in the framework of learning systems.”

Dr Watson’s recent work, which was featured as the front cover story in New Scientist last week, defines the formal links between evolution and learning that enable results to be transferred from computer science to update our understanding of biological evolution. He said: “This work suggests that these feedbacks are not just ‘a complication’ but change the capabilities of Darwinian evolution; specifically, evolution is smarter than we realised.”

The grant, entitled “Putting the extended evolutionary synthesis to the test”, is one of the largest to ever be awarded to evolutionary research. It funds 22 inter-linked projects in total – including theoretical development and empirical experiments – and supports a wide range of additional activities that will promote interaction and collaboration between institutions. The other institutions supported by the Templeton grant are Clark, Indiana and Stanford universities, and the Sante Fe Institute, in the U.S.; Cambridge and St. Andrews universities in Great Britain; and the University of Lund in Sweden. Research on the project will start in September 2016.

A Southampton team from Electronics and Computer Science (ECS) has won the 'Best Innovative Applications Paper' award at this year’s annual AAMAS conference – the international conference on Autonomous Agents and Multiagent Systems.

The team of researchers from the Agents, Interaction and Complexity research group in ECS, have been working on electric vehicle (EV) smart charging mechanisms. These mechanisms will become necessary as an increase in EVs will place significant strains on our electricity distribution infrastructure. Their work mostly focuses on how auction-like market mechanisms can be applied to schedule EV charging intelligently and efficiently. In these mechanisms, drivers effectively place bids on electricity, which indicate their willingness to pay and their individual constraints for charging.

The team’s paper, “Bid2Charge: Market user interface design for electric vehicle charging”, won the Best Innovative Applications Paper award at the AAMAS Conference, held this year in Singapore. In it, the team explored how these complex auction mechanisms can be presented through simple interfaces to non-expert users (ie EV drivers). This is a vital step towards making such mechanisms usable and acceptable for their intended end-users. Through experiments with over 300 non-expert users, they showed that simpler interfaces (which significantly reduce the complexity of the bids that drivers can submit) reduce not only deliberation time for drivers, but actually increase their performance at the same time. This means they choose bids that more closely match their driving patterns, without requesting too much or too little electricity for charging.

The paper also shows how reinforcement learning (a type of machine learning) can be used to predict how people respond to these simplified interfaces. This promises to enable the designers of charging mechanisms (eg EV / chargepoint manufacturers or electricity distribution companies) to test wide ranges of interfaces before starting costly user trials.

AAMAS is the main international conference on intelligent agents, and it is also one of the most prestigious AI conferences. The “Innovative Applications” theme is a special track at the conference for applications of agent-based technologies to commercial and/or public policy domains.

The team’s award-winning paper can be found at http://eprints.soton.ac.uk/387250/

Pictured: Team members Dr Enrico Gerding, Dr Sebastian Stein, Professor Nick Jennings and Dr Avi Rosenfeld being presented with their award

The University of Southampton will showcase its world-leading aerospace research and expertise at Farnborough International Airshow (FI16) this week (Monday 11 to Sunday 17 July).

Based in the Innovation Zone (Hall 3, Stand 28), the University will focus on three main themes: autonomy, sensors and cyber security - interdisciplinary research areas involving academics and researchers from Electronics and Computer Science (ECS). It will promote its strategic research groups for Autonomous Systems and MENSUS (Monitoring of Engineered and Natural Systems Using Sensors) and cyber and security capabilities, including the recently established Cyber Security Academy. Other technology being demonstrated throughout the week includes optical sensors for fuel level monitoring, structural sensing kit to detect and analyse movement (in wings, for example) and remote sensing/diagnostic experiments with Boeing.

A new and exciting initiative this year is the University’s inclusion in the Farnborough flying displays where a Southampton Unmanned Aerial Vehicle (UAV) will beam live photography onto TV screens across the exhibition halls.

Chris Lord, from Research & Innovation Services at the University of Southampton, said: “Aerospace is a hugely important area for Southampton and we plan to showcase our extensive research at this important event. The Airshow provides considerable opportunities to strengthen our industry relationships and present our expertise to the thousands of attendees.”

One of the key activities for the University, is the opportunity to take part in Futures Day (Friday 15 July), when the event opens its doors to over 5,000 younger people. The day allows them to see the career opportunities available in the Aerospace and Defence sectors, and participate in a variety of expert talks, educational activities, competitions and group work, all designed to raise awareness of related industries and careers.

The Southampton exhibition will feature degree courses from ECS and allow students to experience some of its exciting areas of research and to take part in a hands-on activity on how safety systems work in planes. ECS' exciting new Aerospace Electronic Engineering degree will be presented at Farnborough, providing considerable exposure to this new course, both for industry and students. Visitors will also have the opportunity to discover ECS' undergraduate and postgraduate degrees in Cyber Security.

Dr Sarvapali (Gopal) Ramchurn is an Associate Professor in Electronics and Computer Science (ECS). He is an expert in artificial intelligence and member of the Agents, Interaction and Complexity research group. And a seasoned Fantasy Football competitor. This year he thinks he's got the ultimate dream team for the 2016 Premier League Season:

"There’s hardly a moment to catch a breath in this exciting summer of sport," he says. "With Euro 2016 already a distant (and dismal) memory and Olympic fever now sweeping the nation, it’s already time for fans to turn their minds back to dramatic twists and turns of the Premier League.

"And what a season it promises to be. The Manchester clubs look rejuvenated under two of the most sought-after managers in world football, Chelsea are tipped to bounce back after an astonishing fall from grace last season and Leicester are ready to defend their title while entering a historic first Champions League campaign. Will this be Arsene Wenger’s final season at Arsenal? Could Liverpool or Tottenham be the dark horses in an unexpected a title charge?

"Millions of players are readying their fantasy football squads ahead of the big kick-off this weekend, and I’m tapping into a (not-so) secret weapon that will optimise my chances of topping my leagues next May. Together with ECS postgraduate Tim Matthews, I’ve created a predictive system that uses Bayesian Machine Learning techniques and a combinatorial optimisation algorithm to bolster my team and transfer selection.

"The tool has been tested on player data in previous seasons and produced teams that would have consistently ranked in the top one per cent of the Fantasy Premier League.

"The system’s continuously crunching through the numbers and new signings from the summer transfer window, and I’ve got a treat for all the fantasy football players out there this week."

To discover the 15-man squad that Gopal's system recommends for a storming season, visit his blog post for the University's Future Worlds team.

The ORCHID project, a major research project involving the University of Southampton, took top prize in the data and connectivity category at the Collaborate to Innovate Awards this week. Organised by The Engineer – the UK’s longest running engineering publication – the awards were aimed at identifying some of the UK’s most impressive and innovative examples of engineering collaboration.

The ORCHID project tackled one of the key challenges in modern computer science – how humans and intelligent software systems can function together in a seamless and effective manner. The overall aim was to make sense of the volume, variety and constant stream of data that is available today from a myriad of resources such as phones, computers and sensors.

The £10m funded project (£5m from EPSRC with the rest from matched support from project partners) directly trained and employed 50 researchers and PhD students from the universities of Southampton, Oxford and Nottingham, together with industrial partners at BAE Systems, Secure Meters UK Ltd, Rescue Global and the Australian Centre of Field Robotics.

The project spawned 30 follow-on projects worth £15 million. It also established a new multi-disciplinary research community and initiated a range of start-up companies. For example, the researchers developed a device that provides users with a thermal analysis of their house to reduce their energy consumption. This technology was used in the aftermath of the Nepal earthquake to save lives. ORCHID researchers mapped priority areas for water filters around Kathmandu as requested by Rescue Global. By using crowdsourced data, they were able to map settlements across the affected areas and identify aid and search and rescue priorities.

ORCHID co-investigator Dr Sarvapali Ramchurn, from Southampton’s Agents Research Group – the largest research group of its kind in the world – said: “This award is a great recognition of how great collaboration leads to great research outcomes. ORCHID brought together researchers from different disciplines to tackle really difficult challenges. I think the main reason this collaboration was so successful is because researchers learnt to trust each other’s methods and capabilities and we were all highly motivated to have an impact both in terms of producing high quality research terms and solving key real-world challenges.”

Dr Ramchurn and the ORCHID team will receive the award as part of The Engineer’s Collaborate to Innovate Conference, which will be held in Manchester in November.

Researchers from the University of Southampton are to develop software that can monitor the equipment, fuel and energy performance of a ship at sea.

The University is part of the Ship Energy Assessment – Condition Optimisation & Routing Enhancement System (SEA-CORES) consortium, which provides a live model of ship performance on global operations. The development of the software is led by BAE Systems and is sponsored by Innovate UK.

SEA-CORES is able to correlate variables that could affect a ship’s performance, such as energy consumption and different weather conditions. Using genetic algorithms to track and capture the live data, SEA-CORES provides those on board with a greater understanding of the vessel’s capabilities across a wide range of operations.

Researchers from Electronics and Computer Science at the University of Southampton will work on monitoring loads on the ship and applying novel machine learning techniques to a domain that has largely been data poor.

Dr Sarvapali Ramchurn, who is leading the Southampton research group, said: “Unleashing such technologies on the marine sector is likely to have a huge impact. The work we are doing at Southampton in terms of autonomous systems and machine learning will help improve the efficiency of ships and detect potential issues before they cause major damage.”

BAE Systems is developing and testing SEA-CORES on a commercial tanker provided by James Fisher Marine Services. The trial will analyse the vibration and trim performance of the vessel, its hull state and monitor the integrity of the ship’s superstructure.

Chris Courtaux, Head of Engineering and Energy Services at BAE Systems, said: “SEA-CORES is able to consider all of the important components which affect the performance of a vessel during deployment.

“For instance, reducing speed may save fuel but increase the wear to the engine if below its optimum performance. This could in turn increase the maintenance requirements for these vessels and reduce their availability. It is crucial that we continue to analyse what more can be done to maintain these vessels in an efficient manner and increase the number of ships available for the Royal Navy fleet.”

The software connects technologies in delivering fuel and engine optimisation through the use of the BAE Systems’ Ship Energy Assessment System (SEAS), together with big data analysis by using System Information Exploitation (SIE) technology.

SEA-CORES has been developed in response to the increasing complexities of modern warships and the amount of data their systems produce. The technology could transform how the Royal Navy and BAE Systems maintain and support warships in the future by using the genetic algorithms to identify the relationships between a ship’s systems, calculate their different permutations and ultimately recommend a strategy to optimise the vessel’s performance.

The growing cyber security threat posed by the increasing interconnectivity of smart devices is to be examined by a new Roke Manor Research (Roke)/ Royal Academy of Engineering (RAEng) Research Chair based at the University of Southampton.

Professor Vladimiro Sassone, Director of the University’s Cyber Security Academy, will use the prestigious title to build upon effective collaboration with Roke as a key industrial partner around a research agenda that focuses on the evolving Internet of Things.

The Internet of Things is made of the growing network of devices that are connected through the internet, developing innovative applications which may expose society to new cyber threats. Professor Sassone, part of the department of Electronics and Computer Science, has set out to analyse this changing field in both cyber-physical systems and cyber-controlled infrastructures through the Research Chair.

“It is a great honour to be selected to advance this cutting-edge research with the support of Roke Manor Research,” he said. “It is my ambition to develop an enduring partnership between academia and industry. This award also represents a reward for the work I led towards the foundation of the Cyber Security Academy, a pioneering partnership between the University and key industry stakeholders, which we plan to develop as a technopole for cyber security in the south of England.”

Professor Bashir Al-Hashimi, Dean of the Faculty of Physical Sciences and Engineering, said: “It is gratifying that Vladimiro has been recognised through this prestigious Research Chair and this reflects the world-leading research activities that exist within our department of Electronics and Computer Science. The research undertaken by Vladimiro is extremely important to create new knowledge that drives positive impact on both society and the economy.”

Subjects within the University’s Faculty of Physical Sciences and Engineering have been placed in the top 100 amongst the Best Global Universities for 2017, according to rankings published by US News and World Report.

Amongst the publication’s subject rankings, Southampton features in the top 100 for: Computer Science, 95th; Engineering, joint 89th; Physics, 95th; and Space Science, joint 95th – subjects on offer from the Faculty’s departments of Electronics and Computer Science, Physics and Astronomy, and the Optoelectronics Research Centre

The Best Global Universities rankings focus specifically on academic research and reputation, encompassing the top 1,000 institutions spread across 65 countries. US News bases its rankings on 12 indicators which focus on global research reputation through publications, books, conferences, citation impact, and international collaboration.

Overall, the University has placed 104th, climbing six places from joint 110th overall last year. The rankings, now in their third year, also place Southampton 36th in Europe and 10th amongst UK universities listed. Harvard University is ranked number one overall with the University of Oxford first in both Europe and the UK.

Click to find the full list of US News and World Report Best Global Universities 2017.