DSDS 19

Permanent URI for this collection

https://dspace.lib.cranfield.ac.uk/handle/1826/21164

Browse

Now showing 1 - 20 of 42

Open Access
A picture from thousands of words
(Cranfield University, 2020-01-08 15:08) Oishee, Kundu,
The UK government announced its intention to purchase aircraft carriers in July 1998 and in December 2017 the first of the two ships were delivered to the Royal Navy. Newspaper reports on the procurement of the Queen Elizabeth class aircraft carriers have varied in their sentiment over the two decades- sometimes expressing positive, congratulatory emotions of pride and joy and at other times expressing criticism, anger, and despair at project management and outcomes. The image is a normalized plot of sentiment scores associated with 2,688 articles published in eleven British newspapers (broadsheets and tabloids) between 1998 and 2018, where sentiment scores have been calculated using the sentimentr package in R. This research method allows viewing project history at a glance, identifying and investigating critical periods of success and failure, accounting for wider public opinion towards public projects, and extracting greater meaning from textual data with speed and consistency.
Open Access
Absence of Presence
(Cranfield University, 2020-01-10 08:45) Binti zulkifli, Safiah
The tiny the object, the more difficult it is to detect. Failed to detect does not simply signifies the absence of presence. The object may be camouflage and immersed in this huge world - as this image demonstrates! With a booming micro-drones easily available for all walks of life, it is hugely possible for these drones to perform stealthy surveillance being unnoticeable. But, wait! This fact can be avoidable with a specifically design radar for a micro-size target detection. By extracting micro-Doppler signature which is induced by mechanical vibration or rotation from a target – in this case is the flapping wings enable the presence more significant and visible. It is the time to monitor the usage of this presence to avoid illegal and illicit activities for further exploitation. Full control of the latest technology ensure the harmonic and peaceful environment.
Open Access
Acoustic Interface States via Material Variation
(Cranfield University, 2020-01-07 15:05) Pearce, Benjamin
Interface states at the boundary between two phononic crystals can arise when the two crystals possess overlapping bandgaps and differing signs in the imaginary components of their impedance [1]. The suitability of a given pair of crystals when designing an interface system can be inferred from the parity of the supported standing edge states marking the beginning and end of their band gap [2]. A crystal that supports a standing wave with an even parity pressure distribution at the beginning of its band gap will possess a differing imaginary impedance sign to a crystal supporting an odd parity state at the start of its band gap [2]. Previous works [1,2,3] have designed such systems by varying the geometry of a crystal, such variations result in the band gap of the crystal closing and re-opening, leading to two crystals with geometries that place them either side of a band gap closure. This will meet the previously given criteria for exhibiting an interface state [3]. The present study uses a variation of materials within a phononic crystal to demonstrate an interface system, whilst maintaining a consistent geometry between the two crystals. FEM Modelling was performed, using the commercial COMSOL software, to show that variations in the material properties of one of the materials within a phononic crystal results in an analogous process of band gap opening and closing. Two, two component 1D phononic crystals consisting of alternating layers of materials ‘A’ and ‘B’ were then designed and modelled to confirm the parity of their band gap edge states. Material ‘A’ was set as 3 mm thick ABS plastic, whilst material ‘B’ was either water or aluminium with thickness of 0.3 mm. The crystals were numerically modelled to provide the transmission spectra both individually and as a combined interface system revealing the predicted presence of an interface state. Initial experimental confirmation is underway, with a preliminary sample of an ABS/water crystal having been constructed, and its transmission spectrum obtained. Results thus far show the presence of the expected modes and a band gap. Further investigations are now underway using an improved sample so that data may be better compared with modelling.References:1. Xiao, M., Zhang, Z. Q., & Chan, C. T. (2014). Surface impedance and bulk band geometric phases in one-dimensional systems. Physical Review X, 4(2), 1–12. 2. Meng, Y., Wu, X., Zhang, R. Y., Li, X., Hu, P., Ge, L., Wen, W. (2018). Designing topological interface states in phononic crystals based on the full phase diagrams. New Journal of Physics, 20(7). 3. Xiao, M., Ma, G., Yang, Z., Sheng, P., Zhang, Z. Q., & Chan, C. T. (2015). Geometric phase and band inversion in periodic acoustic systems. Nature Physics, 11(3), 240–244.
Open Access
Adaptations to unaccustomed resistance training for women in physically demanding occupations
(Cranfield University, 2020-01-09 15:41) McGuire, Stephen
Improvements in maximal strength may reduce injury risk and improve occupational performance. Women’s naturally lower muscular strength compared to men, can be offset through resistance training (RT), but research into its implementation has been limited. Therefore, the time-course of muscular adaptations is of interest, with evidence suggesting adaptations in size and strength are detectable as early as 3-4 weeks, concomitant with an attenuation of RT induced muscle damage. This evidence offers a reduction in the time commitment of RT that may enhance readiness for physically demanding activity, but functional adaptations have yet to be explored. The purpose of this thesis is to explore measurement techniques, moderators of baseline variation and responses to unaccustomed heavy-load RT in civilian women representative of individuals entering physically demanding occupations. Specifically, high reliability and acceptable validity as a measure of maximal strength was established for the isometric mid-thigh pull, now a UK Armed Forces entrance test, in non-athlete females. Next, the influence of physical characteristics, previous sporting activity, and single nucleotide polymorphisms on variation in strength was assessed. The ongoing final study aims to assess the efficacy of a 4-week heavy compound RT program on improving performance and biochemical adaptations.
Open Access
Additively Manufactured (3DP) thermite structures vs conventionally manufactured equivalents
(Cranfield University, 2020-01-09 10:23) McGee, Christine
Research into additive manufacturing (AM) has been steadily expanding over the past five decades. Where once only polymeric materials could be reliably printed, AM has been adapted to print with a range of materials such as biological, metallic, ceramic and even foodstuffs. The advantages of manufacturing in an additive manner include; a) a layer-by-layer approach allows the creation of architecturally complex structures, b) a reduction in weight, c) lessening of waste and d) the ability to create parts that are otherwise difficult or too costly to produce. Pyrotechnic materials, including thermite, are used in a wide range of commercial and defence applications. However, hazards present during manufacturing and storage have resulted in major accidents around the world, with subsequent loss of life and in some cases loss of public infrastructure. AM, using a dry powder printing technique means that parts can be manufactured on demand, reducing the need for storage of large volumes of fully formed products or mixes, thus increasing the safety over lifetime of a product.The performance of pyrotechnics materials is dependent on a number of properties, including chemical composition, thermodynamic properties and physical form. In combination with composition, architecture could be utilised to understand and control these properties. A bespoke printer capable of additively manufacturing pyrotechnic materials has been constructed with the aim to explore this research area. In this paper, we discuss the development of the AM technique and methodology for the burn test experiments. We conclude with the results from the burning of AM thermite structures and compare their performance with conventionally prepared equivalent thermite examples.
Open Access
Advanced Technologies for the Bonding and De-Bonding of Armour Structures (ArmourBond)
(Cranfield University, 2019-11-19 15:37) Chatzikypraios, Konstantinos
A common multilayer vehicular armour design consists of a combination of adhesively-bonded metallic, ceramic and composite layers. The different properties of each layer contribute in fulfilling the lightweight and the high ballistic property requirements of these structures. Due to their complexity, though, there still are limitations in their performance that are not fully understood. These limitations are associated with a lack in understanding and controlling the propagation of the stress waves, which are induced by the projectile upon the impact event, through the different layers. The stress wave propagation, via reflection and transmission phenomena, is controlled by the acoustic impedance mismatch between the layers. The current epoxide-based armour systems, are incapable of attenuating the stress waves, promoting stress transmission, and exposing the occupants of the vehicles to greatly damaging shocks. Additionally, an increased reflection of the stress waves (in the form of tensile stress) causes the ceramic failure. As a result, the multi-hit performance of the armour is degraded. This PhD study will try to overcome these limitations by focusing on the investigation of novel adhesively-bonded systems, of high energy absorbance capability and improved shock wave attenuation, enhancing the multi-hit capability of the armour. The effect of the material selection and surface treatment on the adhesion and on the mechanical and ballistic performance of the novel structures (using the Split Hopkinson Pressure Bar and the Gas Gun apparatus) will be thoroughly investigated. Further to the above, the controlled debonding of these structures (dismantlable adhesion) will be also investigated, with respect to tailoring the armour against specific threats, fulfilling the “fly light and fight heavy” concept. Finally, the upscaling of the developed materials and structures will take place so that novel armour packs can be tested and validated, according to the STANAG standards.
Open Access
An End to End Hyperspectral Scene Simulator with Alternate Adjacency Models and Its Comparison with CameoSim
(Cranfield University, 2019-11-19 15:37) Zahidi, Usman
In this research we developed a rendering based End to End Hyperspectral scene simulator CHIMES (Cranfield Hyperspectral Image Modelling and Evaluation System), which generates NADIR images of passively illuminated 3-D outdoor scenes in visible and reflective infrared region i.e.360 nm to 2520 nm. CHIMES is capable of generating 3D geometry based on Digital Elevation Model (DEM) data. The atmospheric data is generated by radiative transfer code, MODTRAN, which is used to generate sky-dome environment map that also includes polarisation effect of sky due to Rayleigh scattering. CHIMES generates materials’ optical radiance by employing Metropolis Light Transport (MLT) method to simulate light. It includes render-time scattering phase function, adjacency-effect, sky-dome emitter and sensor modelling. CHIMES also includes adjacency-effect model similar to commercial Hyperspectral Image (HSI) simulator CameoSim, developed by Lockheed Martin, UK. This similarity is premised by the results of rigorous comparison of both simulators under several atmospheric conditions which shows that both simulators are in agreement when executed under same atmospheric and similar rendering configurations. An improved adjacency-effect model is also incorporated which is again compared with CameoSim. Results manifest that CHIMES together with this adjacency model perform better than CameoSim with respect to ground truth data.Keywords: Hyperspectral Imaging, Synthetic Image Generation, Adjacency-Effect, CameoSim.
Open Access
An Introduction to Intelligence Engineering: Operating Beyond the Conventional
(Cranfield University, 2020-01-08 15:32) Svendsen, Adam
Contemporary defence and security enterprise efforts can be viably improved. With an overarching focus on ‘ways’, ‘means’, up and across to realising operational and strategic-ranging ‘ends’, this presentation advances a substantially-structured, multi-scaler ‘Intelligence Engineering’ (IE)-based framework and ‘step-by-step’ toolbox useful for both deployment and employment for a multitude of purposes - essentially whatever is to be accomplished. As the presentation goes on to reveal, the IE framework can contribute towards helping progress several tasks, such as including -but not limited to -the work of interest to several different high-level leaders and policy- to decision-makers and other share-/ stakeholders, end-users, customers or clients. Both ‘situational awareness’ to deeper-ranging ‘contextualisation’ assistance value is offered. The different process ‘steps’ cover diverse areas such as, inter alia: ‘focus/topic selection’ through to the fashioning of ‘signiﬁer node(s)’ for helping in decision-making both now and into the future. Concluding, the presentation highlights that the entire process involved facilitates: (i) greater risk appreciation; and then (ii) subsequent risk management; as well as even advancing (iii) risk engineering to resilience qualities. This is in overall defence and security enterprises and endeavours (however precisely they might be deﬁned or determined) during an era when much uncertainty is encountered.
Open Access
Antibiotic Discovery from the Abyss
(Cranfield University, 2020-01-08 09:20) Stennett, Henry
It is essential that we discover and develop new antibiotics to overcome the problems associated with antibiotic resistance and find treatments for bacteria that don’t respond well to currently available treatment options. Recent years have seen revived interest in culture-dependent methods for discovery - screening rare bacteria from unexplored environments for their ability to inhibit the growth of pathogens. The deep sea is vast, rich in biodiversity, and one of the few ‘pristine’ environments on the planet. The extremophilic bacteria from this niche are likely metabolic innovators that evolved differently to terrestrial species, making them attractive sources of novel natural products.Our aims include (i) characterising the microbiome of deep-sea sponges, which have never been investigated before; (ii) culturing deep-sea bacteria and screening them for antibiotic production; (iii) genome sequencing and mining of producers to delineate the biosynthesis of novel antibiotics. This interdisciplinary project involves microbiology, bioinformatics, and analytical chemistry techniques.A recent screen of bacteria isolated from deep sea sponges found six strains that produce antibiotics under specific culture conditions. The genomes of these bacteria have been sequenced with Illumina and Nanopore technologies, and several active molecules have been purified. Our hit rate (8/487, or 1.6%) is higher than estimated for other environments, indicating that deep-sea sponges are a powerful source of biodiversity. Future work will involve linking natural products to biosynthetic gene clusters and transferring these clusters to heterologous hosts for larger scale production of antibiotics for method-of-action and efficacy studies.
Open Access
Artificially Intelligent Targeting
(Cranfield University, 2020-01-15 15:24) Westlake, Samuel
The aim of this project is the development of new techniques for infrared anti-ship missile seekers. This image illustrates how we are using deep learning to detect, recognise and classify multiple ships. Our algorithm can differentiate between military and civilian vessels, and is even robust against the presence of infrared countermeasures and background clutter.In most cases, training deep learning algorithms requires thousands, if not millions, of carefully labelled examples. This presents a major challenge for the application of deep learning to infrared missile seekers, as the availability of such training data is extremely limited. To over come this, we simulated multiple thermal signatures for ten different ships and used these to synthetically generate a large and realistic data set. This data was then used to train our artificial neural network, and the subsequent model performed successfully on real-world infrared test data.
Open Access
Audio Visual Assessment of CBRN hazards and injuries (AVA)
(Cranfield University, 2020-01-08 09:11) Komianos, Athanasios
In recent years, there has been a significant rise in the uptake by the general public, of audio visual (AV) technology. In part this is driven by the wider adoption of smartphones, social media, broadband internet, computers with in built webcams and wearable devices. Despite the widespread adoption of AV technology, it has seen limited use in CBRN incidents, environmental hazards and emergencies.AVA will systematically measure and evaluate the current communication situation, capture and analyse the information requirements of groups of users, whilst identifying any constraints, consult UK, European and international standards and processes associated with communication during a CBRN incident. It will assess, evaluate and support the development of methods for the secure transmission of real time video over low bandwidth connections, to a web-video player, with the ability to adapt to any screen size, work on any smart device and to provide support to CBRN incident responders and field epidemiologists, where there is a shortage of specialists. The project outputs will help to inform decisions about the use of such new technology for front line needs, enabling remote responders in a disaster/hot zone, where there is limited network connectivity, to live-link with back-office specialists. AVA will develop the standards, procedures and guidelines for the use of this technology for remote risk assessment and decision making.
Open Access
Automated Question Generation for Delphi Studies
(Cranfield University, 2019-11-19 15:41) Sidorenko, Tatjana
Surveys are a relatively easy and low-cost method for gathering a group opinion on a given subject. Despite there being several survey platforms the adaptation of these platforms for complex Delphi studies still leaves much to be desired. ‘Qualtrics Delphi Toolkit’ (QDT) is a tool that is designed to make the Qualtrics survey platform better adapted for Delphi studies.The Delphi method has been devised to aid forecasting and decision-making. It is a method that gauges the wisdom of crowds to estimate a solution or to come up with an accepted definition. A Delphi study consists of a number of rounds. In the first round, a panel of experts is gathered, each expert shares their opinion on the selected topic. In the second round, the opinions are compiled into a single unified summary that is then presented back to the experts. Each expert is then given an option to alter their original answer for the changes to then be propagated into the summary answer. This procedure is repeated until a consensus is reached. As the use of the Internet has become widespread it has allowed an increase in the magnitude of such studies, through facilitating access to a greater range of experts, as well as enabling the rapid gathering of more information. The tool support, however is often not able to efficiently manage this scale of data collection. Certain functionality, such as the ability to randomly sample a set of questions is not efficiently achievable at the scale required for large Delphi studies. The need for better tool support has become evident after facing the challenge of presenting a list of over 300 hacking techniques to a group of experts. This required stochastically presenting a subset of these techniques maximising expert engagement whilst minimising fatigue and performance decline.QDT is a tool that is written to extend the functionality of a well-known survey platform, Qualtrics, to automatically generate questions relating to a Delphi study. Currently, Qualtrics offers a feature called ‘Piped text’ that passes information from one question to the next. However this feature alone is not enough when there is a requirement to generate large volumes of questions based on previous questions. QDT is designed to automatically create questions from a list of strings. Together with Qualtrics's conditional formatting it allows efficient creation of large volumes of questions where the question text depends on user's previous choice. This will enable the efficient creation of large-scale survey-assisted Delphi studies where the list of topics is very large.
Open Access
Bone Disease: The Integration of Nano- and Macro-Scale Studies
(Cranfield University, 2020-01-09 11:44) Arnold, Emily
Bone is a hierarchical structure that provides support and protection to a body. In the macro-structure of bone, geometry affects it ability to perform these functions, while in the micro-structure the material properties dictate changes. The easy substitution of atoms into the crystal, as well as the binding of extraneous proteins to the surface, have been found to alter many material properties.This research aims to develop a greater understanding of the fundamental properties and behaviours of the mineral component of bone (biological Hydroxyapatite or HA) through several techniques, some that are well documented within the field of research and some that are not. While X-Ray Diffraction (XRD) and Fourier Transform Infrared Spectroscopy (FTIR) are often used in the study of HAp, this research aims to expand the lesser-used methods used to include X-Ray PDF Studies with the use of Synchrotron Radiation, as well as Differential Scanning Calorimetry (DSC), and Thermogravimetric Analysis (TGA).By building a cohesive model one step at a time we can begin to understand fundamental material characteristics of this fascinating building block of life.
Open Access
Breast microcalcifications as biomarkers of tissue pathology
(Cranfield University, 2020-01-09 11:31) Gosling, Sarah
Breast cancer is the most commonly diagnosed cancer in women in the UK, accounting for 31% of all cases. Some breast malignancies, known as in-situ cancers, have the potential to form invasive cancer, but this is not true in all cases. There is significant overtreatment of in-situ cancers, which can be a traumatic process for patients. Therefore, a robust method to determine which of these malignancies will develop into clinically significant invasive cancer is required. Ductal carcinoma in-situ (DCIS), the most common in-situ breast cancer, has associated calcium deposits (microcalcifications) in 80-90% of cases. Microcalcifications are therefore an important diagnostic indicator of DCIS. These microcalcifications are composed of hydroxyapatite, a nano-crystalline calcium phosphate, with the potential for high levels of ion substitution.The crystalline nature of hydroxyapatite permits the use of X-ray diffraction to investigate the microstructure of microcalcifications found in breast tissue. X-ray diffraction patterns can be used to determine properties such as crystallite size, irregularities in the crystal lattice and give indications about the ions present. Together, differences in these parameters may permit the development of a novel prognostic marker for different breast tissue pathologies.
Open Access
Building Management Systems: Context, Collaboration and Organisational Security
(Cranfield University, 2020-01-09 09:37) Boakes, Emma
Physical assets are increasingly connected to organisations’ networks, exposing organisations to the risk of cyber threats that could have a physical impact. There are several instances where this has already happened, and industry reports show that the number of groups interested in targeting such systems, and the number of cyber incidents on physical systems, is increasing. A Building Management Systems (BMS) is one such connected system. A cyber-attack targeting a BMS could disrupt heating, ventilation or power, or could undermine physical security operations which increasingly rely on internet-enabled devices, such as CCTV cameras and access control. In addition, these systems could be used as a gateway into the organisation’s network if not properly secured. It is increasingly important to consider a BMS within the context of an organisation’s overall protective security posture. Inclusion of a BMS on an organisation’s network not only increases the attack surface but also places a greater load on security staff, potentially leading to vulnerabilities. Additionally, a BMS spans the boundaries of cyber and physical security, so cyber, personnel and physical security teams need to work together to identify, understand and mitigate vulnerabilities.This research argues that current guidance fails to address the challenges of integrating cyber, personnel and physical security to protect a BMS. Guidance focuses on securing the attack surface with technological solutions and a ‘defence in depth’ approach but overlooks the impact this will have on staff managing the system. Increased system complexity puts yet more load on staff, further reducing their capacity, which in itself could create further vulnerabilities.Furthermore, guidance does not adequately address how separate security teams should collaborate to ensure effective solutions are implemented. The formal collaboration between security teams through a converged security approach has, however, been advocated by industry practitioners and the benefits of such an approach have been highlighted. Despite this, there is little evidence for the approach, and little detail about the type or level of convergence required to achieve the specified benefits. Moreover, there is no indication of how organisations might overcome the challenges of adopting such an approach.This research seeks to explore the context in which a BMS is implemented to inform organisations of the broader system implications around converged security. Initial research explores how separate security teams within organisations collaborate or converge. This presentation will outline the preliminary findings and highlight areas where future research is needed.
Open Access
Buried object detection and classification using SAR in varying moisture environments
(Cranfield University, 2020-01-08 10:18) Sabiers, Richard
Remote investigation and classification of buried objects is important for many reasons. For defence roles the identification of suspicious objects can enhance security. Civilian applications are also applicable allowing estimation of crop yield and remote monitoring of plant health. The aim of this work is to demonstrate a prototype open-source system of radar-based target detection and classification. Experimental targets consist of buried artefacts, including an example of military ordnance, such as a landmine and a metallic improvised explosive device. These represent real world examples, chosen due to differing composition of materials, and will be imaged when buried in a medium of top soil with different moisture levels.Detection of targets is a three-stage pipeline of data collection, image formation and classification. Data is collected using Cranfield University’s prototype mini-GBSAR system, which is deployable in field. This tool enables sub-surface sensing by measuring the backscatter of electromagnetic waves in the microwave region. Synthetic aperture radar signal processing is used to produce the final three-dimensional image. Image formation is beneficial for an intelligent machine classifier to evaluate the data and to separate objects as targets of interest from irrelevant clutter.
Open Access
Buying Butter and Guns
(Cranfield University, 2020-01-07 14:35) Oishee, Kundu,
Is defence procurement different from other procurement activities of the government? This poster will present four cases of recent public procurement in the UK and compare the projects from their inception (tendering, selection of supplier) to final delivery in order to find the similarities and differences in military and civilian public procurement projects.It is easy to assume that defence is different or somehow ‘peculiar’ by claiming that the market is uncompetitive or that it is excessively political. However, there are rich literatures in management of major projects, innovation in large technical systems, and the economics of procurement in imperfect markets which suggest the potential for shared learning from seemingly unique projects in order to improve public procurement performance (i.e. to ensure projects are completed on-time and within budget).Methods and dataThe research project uses comparative case study to answer the research question and examines the procurement of the Queen Elizabeth class aircraft carriers, Ajax armoured fighting vehicle and the procurement of high-speed trains for the Thameslink and the Intercity Express Programme. Data is being obtained from government documents and reports (National Audit Office, Parliamentary Select Committees, government departments), Parliamentary debates, newspapers, trade press/industrial magazines, and interviews. The poster will present the data collected in a format that can easily inform the audience about the similarities and differences that can be observed between these four projects along three dimensions:(1) Competition: points related to the market structure, the level of competition between suppliers, and the selection of supplier(2) Challenges: technological, contractual, and political risks encountered in project delivery(3) Management and governance: scope and structure of project teams, training and capacity, turnover, performance review and accountabilityFindings and conclusionThe research uses multi-disciplinary perspectives (economics, management, public policy) for understanding procurement. The findings will be of interest to those interested in military procurement as well as those interested in the potential for cross-learning between the military and non-military sector on the topic of procuring large technical systems.Although data collection is underway and will be completed only by October, comparing the procurement of the Queen Elizabeth class aircraft carriers and rolling stock for the Thameslink programme already reveals important similarities in competition and challenges faced in the procurement of large technical systems by both the Ministry of Defence and the Department for Transport. Case study research is often criticized for not being generalizable but finding similarities in procurement experience for extremely complex projects (as the ones chosen) casts doubt on the assumption that defence procurement is different from civilian public procurement.
Open Access
Characterising bremsstrahlung and non-linear Compton scatter in high intensity laser-foil interactions
(Cranfield University, 2020-01-08 09:53) Morris, Stuart
Non-linear Compton scatter (NCS) and bremsstrahlung radiation both contribute to the X-rays produced in laser-solid interactions. While it is expected that NCS will efficiently generate X-rays for next generation lasers, the weaker NCS yields from laser-solid interactions in existing facilities have not yet been observed due to the bremsstrahlung background. This work attempts to derive a quantitative description of the bremsstrahlung efficiency, for interactions between ultrahigh-intensity short-pulse lasers and thin-foil targets. This model was tested against simulations from the particle-in-cell code EPOCH, which has been extended to include bremsstrahlung modelling capabilities. The relative efficiencies of bremsstrahlung and NCS will be compared to seek NCS signatures for experimental observation.
Open Access
Deceptive Autonomous Agents
(Cranfield University, 2020-01-09 10:31) Sarkadi, Stefan
Recent advances in Artificial Intelligence (AI) along with recent events revolving around the problem of fake news indicate new and critical potential threats to intelligence analysis, defence, security, and, by extension, to modern society in general. One such threat that we can derive from the development of AI is the emergence of malicious autonomous artificial agents that could develop their own reasons and strategies to act dishonestly. In order to be able to prevent or mitigate the malicious behaviour of deceptive artificial and autonomous agents, we must first understand how they might be designed, modelled, or engineered. In this work, we aim to model and study how artificial agents that deceive and detect deception can be engineered, as well as how such agents might impact the common good.
Open Access
Deep Learning Techniques for Missile Seeker Automatic Target Recognition
(Cranfield University, 2020-01-15 15:30) Westlake, Samuel
Modern infrared missiles use sophisticated computer vision techniques, in conjunction with imaging seekers, to automatically detect and localise targets. However, simulations show that soft-kill countermeasures remain an effective defence against such systems. This research explores the feasibility of deep learning algorithms for Automatic Target Recognition (ATR) and aim to significantly improve seeker performance in the presence of soft-kill countermeasures and clutter. State-of-the-art neural network architectures were benchmarked using both simulated and real-world infrared data. Their performance was also analysed to inform tailored and novel developments. In addition to this improvement to existing capabilities, these algorithms established additional capabilities of target recognition and identification. This effectively enables target prioritisation and safeguarding of friendly assets