Robotics is changing the landscape of innovation. But traditional design approaches are not suited to novel or unknown habitats and contexts, for instance: robot colonies for ore mining, exploring or developing other planets or asteroids, or robot swarms for monitoring extreme environments on Earth. New design methodologies are needed that support optimising robot behaviour under different conditions for different purposes. It is accepted that behaviour is determined by a combination of the body (morphology, hardware) and the mind (controller, software). Embodied AI and morphological computing have made major progress in engineering artificial agents (i.e., robots) by focusing on the links between morphology and intelligence of natural agents (i.e., animals). While such a holistic body-mind approach has been hailed for its merits, we still lack an actual pathway to achieve this. While this goal is ambitious, it is achievable by introducing a unique methodology: a hybridisation of the physical evolutionary system with a virtual one. On the one hand, it is appreciated that an effective design methodology requires the use and testing of physical robots. This is because simulations are prone to hidden biases, errors and simplifications in the underlying models. Simulating populations of robots (rather than just simulating specific parts) leads to accumulated errors and a lack of physical plausibility: the evolved designs will not work in the real system. This is the notorious reality gap of evolutionary robotics. On the other hand, evolving everything in hardware is time and resource consuming. One of our major innovations is to run simulated evolution concurrently with the physical and hybridise them by cross-breeding, where a physical and a virtual robot can parent a child that may be born in the real world, in the virtual world or in both. The advantages of such a hybrid system are significant. Physical evolution is accelerated by the virtual component that can run faster to find good robot features with less time and resources; simulated evolution benefits from the influx of genes that are tested favourably in the real world. Furthermore, monitoring of and feedback from the physical system can improve the simulator, reducing the reality gap.

This project addresses the problem of "characterisation" of Extreme Environments (EE), by deploying and combining information from a variety of different Remote Sensing modalities. Our principle application area is nuclear decommissioning, however our research outputs will be relevant to other EE. Before nuclear decommissioning interventions can happen, the facility/plant being decommissioned must be "characterised", to understand: physical layout and 3D geometry; structural integrity; contents including particular objects of interest (e.g. fuel rod debris). 3D plant models must further be annotated with additional sensed data: thermal information; types/levels/locations of contamination (radiological, chemical etc.). Characterisation may be needed before, during or after POCO (Post Operation Clean Out). "Quiescent buildings" may be over half a century old, with uncertain internal layout and contents. Characterisation is needed in dry environments (e.g. contaminated concrete "caves") and wet environments (e.g. legacy storage ponds). Caves may be unlit, causing difficult vision problems (shadows, contrast, saturation) with robot-mounted spotlights. Underwater environments cause significant visibility degradation for RGB cameras, and render most depth/range sensors unusable. New technologies, e.g. acoustic cameras, engender interesting new challenges in developing algorithms to process these new kinds of image data. In many cases, robots are needed to deploy Remote Sensors into Extreme Environments and move them to desired locations and viewing poses. In some cases, robots must also assist characterisation by retrieving samples of contaminated materials. In many case real-time Remote Sensing data must also be applied to inform and control the actions of robots, while performing remote intervention tasks in EE. This project brings together a unique, cross-disciplinary and international team of researchers and institutes, spanning three continents, to address these challenges. End-users NNL and JAEA will advise on scenarios and challenges for Remote Sensing in nuclear environments. Active facilities at JPL will be used to measure degradation of sensors, chips and software under a variety of radiation types and doses. JPL and Essex researchers will use this data to develop new models for predicting such degradation. Essex researchers will then develop new methods for software and embedded hardware design, which overcome radiation damage by incorporating new approaches to fault detection, tolerance and recovery. The scenarios provided by the partners, and the degradation data measured by JPL, will be used to develop new benchmark data-sets comprising data from multiple sensing modalities (RGB cameras, depth/range cameras, IR thermal imaging, underwater acoustic imaging), featuring a vairiety of nuclear scenes and objects. UoB and Essex researchers will develop new algorithms for real-time 3D characterisation of scenes, with intelligent and adaptive fusion of multiple sensing modalities. First, new multi-sensor fusion methods will be developed for 3D modelling, semantic/meta-data labelling, recognition and understanding of scenes and objects. Second, these methods will be extended to incorporate new algorithms for overcoming extreme noise and other kinds of degradation in images and sensor data. Third, we will develop the robots and robot control methods needed to: i) deploy remote sensors into extreme environments; ii) exploit remote sensor data to guide robotic interventions and actions in these environments. Finally, we will carry out experimental deployments of these new technologies. Robust hardware and software solutions, developed by Essex, will be tested in active radiation environments at JPL. We will also carry out experimental robotic deployments of sensor payloads into inactive but plant-representative nuclear environments at NNL Workington and the Naraha Fukushima mock-up testing facilities in Japan.

The Earth's atmosphere and oceans are warming as a result of increased concentrations of greenhouse gases. Glaciers melt when the Earth warms and water that was stored as ice on land runs off into the ocean and increases sea-level. Over the last few decades, measurements have shown that sea-level is increasing by around 3 millimetres per year, and that this is due to the expansion of the warmer ocean water and the runoff from glaciers. In the 20th century, the sea-level contribution from melting ice was dominated by small mountain glaciers and ice caps, but it is now known that the vast ice sheets in Greenland and Antarctica are contributing an equally large amount to sea level and that their contribution is accelerating. One of the main ways in which ice sheets contribute to sea-level (especially in Antarctica, but also in Greenland) is through rapidly-flowing outlet glaciers that transfer ice from the interior to the margins, where it breaks off as icebergs. Recent measurements, mostly using observations from satellites, have shown that many outlet glaciers are thinning and retreating and, in some cases, their flow is also accelerating. This helps explain why their ice discharge is increasing. These changes in outlet glaciers are complex, but scientists think that they are caused by warmer ocean temperatures and, in some cases, by the landscapes underneath the outlet glaciers, especially if they flow through deep valleys that are below sea level and get deeper inland under the ice. The most dramatic changes have been observed in Greenland and West Antarctica, which store around 6 and 4 m of sea-level equivalent, respectively. Thus, unlike smaller mountain glaciers, changes in outlet glaciers could contribute several metres to global sea-level, possibly over quite short time-scales (just a few centuries according to some predictions). It is for this reason that a lot of research is aimed at monitoring outlet glaciers in Greenland and West Antarctica. Most of the ice in Antarctica is, however, stored in East Antarctica, which holds a sea-level equivalent of around 53 m. It is perhaps surprising, therefore, that there are so few measurements of outlet glaciers in the East Antarctic Ice Sheet (EAIS), but this is probably because it was traditionally thought to be much more stable than West Antarctica. Recently, however, evidence has been uncovered which indicates that parts of the EAIS, especially those parts that that overlie deep valleys and basins, might have retreated quite dramatically when climate was slightly warmer in the past. Moreover, observations of just one or two glaciers in these same regions indicates that they are also thinning and retreating, similar to those in Greenland and West Antarctica. Thus, there is a small but growing body of evidence suggesting that some parts of the EAIS might also be vulnerable to global warming. Unfortunately, we do not have enough observations to know exactly what is happening in different parts of East Antarctica and there is a large amount of uncertainty about whether its outlet glaciers are sensitive to changes in the ocean and/or atmosphere. This project has been designed to specifically address this uncertainty. We will use satellite measurements to determine recent changes on some of the largest and most important outlet glaciers from different regions of East Antarctica. This will tell us where the most dramatic changes have taken place and which areas are more stable. We will then use a computer model to see what kind of changes would take place if air or ocean temperatures increase in the future. This will tell us which glaciers are most sensitive and what their contribution to sea level might be over the next few centuries. Even where glaciers are currently stable, it is important to know by how much climate would need to change before they might react. This new knowledge is vitally important to help governments plan for future changes in sea-level.

Despite the well-recognised influence of clouds and precipitation on our climate, there are still critical gaps in our ability to observe cloud properties that are needed to test and improve how cloud processes are represented in models. This leads to clouds and aerosols being the biggest source of uncertainty in climate models, according to the IPCC. In addition, uncertainties about cloud processes have important impacts on our ability to predict the weather, because precipitation is produced by clouds, clouds modulate the amount of sunlight we receive during the day and heat we lose at night, and latent heat processes in clouds and precipitation drive dynamical changes in storms. Low-altitude clouds of liquid water droplets cover large swathes of the globe, and cool the earth's climate. However our ability to simulate these clouds in climate models is poor, and the production of drizzle has been identified as a key weakness. We need new observations to unravel the processes in these clouds and improve their representation in simulations. Meanwhile ice clouds cover around one third of the earth at any one time, and provide a net warming on average. However the magnitude of this warming is very uncertain, and their impact on our climate is very sensitive to what we assume about their physics. Thus we urgently need to constrain those physical processes controlling how ice particles evolve in natural clouds. Finally, stratiform precipitation is an important component of the hydrological cycle and the radiation budget. Typically such precipitation include an ice phase aloft and a liquid phase at lower altitude. Yet there are processes in both phases which remain uncertain, and require new observations to robustly constrain them. Our novel proposal exploits new radar technology to break through the current limitations on the information we can currently retrieve about cloud properties and the processes that drive the evolution of the hydrometeors within them. With the help of our project partners at the Met Office and the ECMWF we will use this information to improve the simulation of cloud processes in weather and climate forecasts. In 2018 the UK Space Agency and Centre for Earth Observation Instrumentation agreed to fund the development of a new 200 GHz (G-band) Doppler radar system, called GRaCE, led by investigators Huggard and Battaglia. This ground-breaking demonstrator instrument will collect its first data at the Chilbolton Observatory early in 2020, and will be able to penetrate multiple layers of clouds with unprecedented sensitivity to small sub-millimetre particle thanks to the radar 1.5 mm wavelength, the smallest for any cloud radar system worldwide. The radar will be operated for 22 months in synergy with a suite of other remote sensing instruments. The unprecedented dataset will be exploited by GRACES scientists who are leaders in radar remote sensing techniques and have spearheaded retrieval techniques for multi-wavelength Doppler radars. Vertical profiles of cloud physical properties including water content as well as drizzle drop and ice crystal size distributions will be obtained and this data will be used to test the representation of cloud processes in numerical models in much greater detail than has been possible before. Through this leap forward in our ability to observe clouds the GRACES system will become the forerunner for future development of a new stream of ground-based remote sensing instruments, greatly strengthening the current Earth observing system. The high frequency of the radar means that it will also be suitable for development into air-borne/space-borne instruments for cloud related studies, and indeed the proposal is very timely given parallel efforts at NASA's JPL to build an airborne differential absorption radar (for measuring water vapour) at smaller frequencies (165 to 173 GHz), and to develop CubeSat radars in the G-band (see NASA-JPL's LoS).

Aeolian (wind-blown) sand dunes occupy 10% of the Earth's surface, both in vast desert sand seas and as important natural defences against flooding along coasts. While the environmental conditions that influence the shape, movement and patterns of fully grown dunes have been extensively studied, arguably the most enduring deficiency in our understanding of these landforms is also the most profound: how do wind-blown dunes initiate? Initiation is central to understanding dunes as major geological units, including the response of these landscapes to climatic drivers, environmental change and societal impact. The significance of dune initiation for the wider understanding of wind-blown sandy systems and their contexts, for which the discovery of extra-terrestrial dune fields has added a recent impetus, ensures that the question of initiation has remained prominent throughout the history of desert research. Despite this, existing ideas proposed to explain processes of dune origin have remained largely descriptive and uncorroborated. The persistence of the question regarding dune initiation is not due to an absence of appreciation of its importance but, rather, a lack of the means to tackle this fundamental issue. The critical obstacle to a fully developed understanding of dune initiation is that, until now, measurement of the necessary variables, at the ultra-high spatial and temporal resolutions required to detect small-scale variations in surface conditions and wind-blown sand transport, has been impossible. Recent technological advances in the geosciences both inspire and underpin this proposal, as they now provide the opportunity to meet the demanding requirements of process measurement. Surmounting the abiding problem of dune initiation requires novel approaches in research design and our proposal tackles the issues of measurement at small scales by forging complementary links between fieldwork and physical modelling, as well as an ability to widen the application of detailed process findings through computer modelling. Specifically, this proposal will for the first time examine the key inter-relationships between airflow, surface properties, changes in sand transport and bedform shape that lie behind a meaningful understanding of how nascent dunes emerge. Full measurement of controlling processes and bedform development will be achieved through field monitoring of surface properties and bedform change at extremely high resolution. A key novelty of the fieldwork is that it will be carried out at three carefully chosen locations of known dune development, with each location representing the 'type site' for three different drivers of dune initiation; surface roughness, surface moisture and sand bed instability. The fieldwork will inform experiments undertaken in a bespoke laboratory flume that is designed to enable accurate characterisation of flow very close to the 3D surface of modelled dunes using state-of-the-art imaging techniques. Our field and laboratory dataset will be used to drive a computer model that we will then run to test the sensitivity of dune initiation and growth to different controls in a range of environmental conditions in deserts, coasts and on other planets. Our proposal is built on a new capability to make field observations at the requisite exceptional levels of detail, augmented by closely coupled state-of-the-art laboratory flow simulations, plus the development and application of evidence-based modelling to examine drivers of dune initiation. In concert, this approach represents an unprecedented opportunity to overcome a truly enduring plateau for understanding the origins of one of the major terrestrial landform systems.