Theme:

Human-Computer Interaction

Funding:

EPSRC

Summary on Grant Application Form More than two thirds of the worlds population will be living in cities by 2050, which made sustainable cities a key theme at the Rio+20 Summit. Over 40% of the UK population live their working lives carrying out knowledge work in offices - enclosed spaces in the built environment. Over 40% of the UK's GDP is driven by the knowledge work carried out in these buildings. While these buildings must meet basic environmental and safety standards, building codes are primarily defined around the performance of the physical infrastructure rather than capturing the performance of a person in a building in a city. Indeed, the energy agenda has recently focused on making these work environments as air tight and thus as energy efficient as possible. But what if this energy efficiency is costing our wellbeing, and thus our performance and capacity to contribute to environmental, economic and social sustainability? We are all familiar with working in buildings that are overhot, stuffy and have seemingly no air flow, and how our performance seems to suffer as a result. Yet such environments may be well within building specifications for environmental quality. There is a clear need to be able to quantify the impact of indoor air on human performance and determine what is optimal for quality performance.

We hypothesize that poor indoor environments are the result of approaching building performance on the basis of (1) what is easy to measure (e.g. energy/carbon) and (2) without properly considering effects of the environment on human cognitive performance. Poor design may have such a critical impact on the creativity and innovation required for knowledge work that we need a radical shift in design focus. That radical reconsideration may show that even small changes not just to design, but to the way information about both the building and the person is presented, may significantly improve performance.

The goals of the Refresh project are to put the human at the centre of building performance and to develop new measures and models that better capture the complexity of these interactions. We plan to measure the *dynamic* changes in and around the local environment on human performance in that environment. These measures will give us ways to create new models of building environments that in turn will be available to help inform policy for building quality that takes human wellbeing in these environments into account. These models will also help us to design new kinds of environmental interaction tools. For instance, much current attention is focused on smart meters to encourage us to change our behaviour around energy use. This is a very one-way view of our interaction with a buiding; it's all building to human. What if the building, however, knew something about our state and tasks? Would we be able to present a co-interaction meter that might suggest opening a window or going for a walk to get some air in order to complete a task when we're apparently becoming sluggish?

Our proposal brings together a novel mix of ICT and engineering sciences. As such, our results will affect a range of disciplines. Overall, our research explicitly aims to develop a methodology for assessing indoor climate for human performance as the outdoor microclimate is altered due to a changing neighbourhood. We will discover how dynamic changes may enable us to create not just a sustainable indoor environment, but an indoor environment which sustains. We will enhance the present energy-dominated portfolio of built environment research by taking a holistic view and integrating explicit feedbacks between urban microclimate, ventilation and human performance. Our approach will provide guidance on how to meet energy efficiency targets without compromising productivity.

Primary investigators

• monica schraefel
• Janet Barlow
• Catherine Noakes

Associated research group

• Agents, Interaction and Complexity

Date:

2013-

Theme:

Open & Linked Data

Funding:

EUFP7

Associated research group

• Web and Internet Science

Date:

2014-2014

Wind is one of the significant renewable energy resources, which has been highly promoted. However, the biggest challenge of wind development is the unreliability due to the intermittence. One of the solutions is hibernus restore, which allows electronic applications working in dynamic and unreliable power situation by storing current data and states in somewhere else before wind power is actually getting lost. Therefore, accurate prediction of wind power is required. The objective of this project is to implement the system, which can predict the change of wind power in very short-term (few seconds) and foresight if wind is about to lose (below the threshold). Several prediction algorithms are evaluated first in MATLAB, which contain persistence method, ARIMA and BPNN. These approaches would be compared with appropriate error metrics to figure out the one with the best performance to be finally implemented in FPGA board. The evaluation indicates BPNN performs best to predict wind in very short-term. In order to implement BPNN, shift-and-add PWL approximation method is applied to approximate the transfer function of BPNN. Finally, used energy, resources and calculation speed of the system would be used to evaluate the final performance of the ANN system.

Primary investigators

• Alex Weddell
• Geoff Merrett

Associated research group

• Electronics and Computer Science

Date:

2014-2018

Theme:

Simulation, Modelling and Evalution

Unlike the established Sub-Threshold Voltage (STV) techniques, the recently introduced Near-Threshold Voltage (NTV) approach promises to scale well with the decreasing supply voltage while maintaining robustness and reliability. This research proposes to integrate the NTV approach, as a method of enabling significant reductions in the energy consumption, with many-core processor architectures. One of the main goals of the proposed research is to investigate to what extent extremely low supply voltages can maintain the robustness of logic circuits, tolerate process variability and be resilient against errors. Since NTV circuits at low voltage must run at lower frequencies, CPU intensive tasks, that are typically performed in battery powered mobile devices, will be executed in a many-core environment to maintain performance while the overall energy is reduced. As natural variations in manufacturing processes can lead to voltage fluctuations and logic circuit failures, techniques will be investigated to monitor the errors and adjust the supply voltage dynamically to maintain reliable operation. Control circuitry will need to be developed to reduce energy consumption even further, through adjusting the required supply voltage of individual components of the chip and powering them down when they are not being used. It is envisaged that processors developed in the course of this research may also be suitable for other applications, for example, energy-harvester powered sensor nodes where the development of specialised ultra-low energy processor technologies is currently lagging.

Associated research group

• Electronic and Software Systems

Date:

2010-2014

Funding:

EPSRC

A normal functioning hand is one of the most important features for human independence. A complex neuromusculoskeletal structure of the human hand consists of many dedicated subsystems cooperated in a highly organised manner to form a powerful and precise device. The malfunction of any of its elements may result in disability and hand functional impairment. There exist many injuries that can result in the loss of hand function, such as i.e. stroke, spinal cord injuries and complications after hand surgery or hand traumatic injury such as i.e. tendon ruptures.

Surface electrode array stimulation is a non-invasive method of muscle activation applied via adhesive electrode arrays placed on the surface of the patient skin above the location of the desired muscles. Surface electrode array stimulation is able to induce movements in paralysed or weak limb, by delivering a series of electrical pulses to associated skeletal muscles through activation of chosen elements of electrode array. Surface electrode array stimulation is a promising technique for stroke rehabilitation of the wrist and fingers, due to its increased muscle selectivity. However, the effectiveness of the method is strongly related to the precision and accuracy of the stimulation. Locating the optimal stimulation sites via selection of the appripriate elements of the electrode array is critical to the effective application of this rehabilitation approach.

This project addresses the use of surface electrode arrays to regulate the stimulation applied to the extensor muscles of the hand and wrist in order to induce hand movement to desired posture. A general control strategy developed in this project embeds optimisation methods for selection of appropriate elements of the electrode array with iterative learning control.

In iterative learning control, the patient makes repeated attempts to complete a predefined task with the aim of gradually decreasing the error between the movement performed and desired one. A number of different gradient-based methods, such as penalty method and sparse optimisation methods has been developed and tested based on theoretical and experimental findings. These methods are used to find a sparse input vector, which is employed to select only those array elements that are critical to task completion within iterative learning control framework.

The developed methods, are presented in the context of the complete and novel design of the Hand Rehabilitation System (HaReS) to provide both, theoretical and practical indications for further expantion of this rehabilitation technology and future research. The system comprises the ILC-based control algorithms for electrode array stimulation with a game-based training environment that provides feedback to the patient.

Primary investigators

• Anna Soska
• Eric Rogers
• Christopher Freeman

Associated research groups

• Electronics and Electrical Engineering
• Electronics and Electrical Engineering
• Electronics and Electrical Engineering