7103 research outputs found
Sort by
X-ray computed tomography and digital volume correlation insight into microscale mechanisms of rock fracture and fatigue
No full textDeep learning : assisted visualisation of cellular structure in microscopy
No full textWithin a few years, deep learning (DL) has become an important staple in image processing in microscopy, often showing superior performance in comparison to classical methods. Specifically, the ability of an artificial neural network (ANN) to learn a complex mapping function between image domains has found wide application. Its utilisation in fields such as image restoration, image segmentation, and artificial labelling allows cell biologists to increase knowledge output from their microscopy data. However, the performance of an ANN for such tasks is limited by the quality of the data that is used for the training. In practice, high quality ground truth data can be difficult to acquire with microscopy due to the multitude of limitations of imaging setups as well as labelling protocols. Alternatively, target data for the training can be generated through manual annotation but this is often unfeasible as it is extremely time- and labour-intensive, and requires a high level of expert knowledge. In this thesis, methods are introduced for image-to-image translation that were developed while taking those limitations into account, with the main aim of enhancing the microscopic visualisation of routinely studied cellular structures in cell biology such as the actin cytoskeleton and the microtubule network. Two methods are presented. The first, Label2Label (L2L), is a new restoration method in fluorescence microscopy that enhances the image contrast of cellular structure. L2L does not require clean target data for the training. Instead, clear differences in image quality observable between non-identical fluorescent labels are capitalised on to systematically train a network to increase structural contrast. The idea to dual-label a cell to acquire training data makes L2L relatively straightforward to implement in practice.
Moreover, results show that a multi-scale structural similarity loss function can enhance the performance in L2L when implemented carefully. Furthermore, the use of artificial labelling in interference reflection microscopy (IRM) is explored. In artificial labelling, an ANN is trained to translate between images acquired with the label-free IRM technique and fluorescence images. Both images differ in their specificity; contrast in label-free images stems from all cell components, whereas fluorescence images are ideally highly specific to the cellular target that was labelled. Consequently, with artificial labelling, specific cellular structures can be selectively visualised in label-free images. For the translation between IRM images and fluorescence images of a focal adhesion-marker, different cases were explored where target data exhibited either a low or a high contrast of the focal adhesions, and were paired or unpaired to the IRM inputs. Results presented in this work show that a network can be successfully trained for artificial labelling in IRM. Moreover, high quality paired target images are not necessarily required for that task. Here, a newly developed framework, the so-called 2LGAN, that is a variant of a generative adversarial network, performed highest for such a case. Notably, L2L and artificial labelling in IRM out-perform classical methods as pre-processing step for downstream quantitative image analysis. Both methods could have important impact in the future for the visualisation of other cellular structures not explored in this work. One such structure could be the plasma membrane which is difficult to visualise with microscopy but is of high interest in cell biology.Within a few years, deep learning (DL) has become an important staple in image processing in microscopy, often showing superior performance in comparison to classical methods. Specifically, the ability of an artificial neural network (ANN) to learn a complex mapping function between image domains has found wide application. Its utilisation in fields such as image restoration, image segmentation, and artificial labelling allows cell biologists to increase knowledge output from their microscopy data. However, the performance of an ANN for such tasks is limited by the quality of the data that is used for the training. In practice, high quality ground truth data can be difficult to acquire with microscopy due to the multitude of limitations of imaging setups as well as labelling protocols. Alternatively, target data for the training can be generated through manual annotation but this is often unfeasible as it is extremely time- and labour-intensive, and requires a high level of expert knowledge. In this thesis, methods are introduced for image-to-image translation that were developed while taking those limitations into account, with the main aim of enhancing the microscopic visualisation of routinely studied cellular structures in cell biology such as the actin cytoskeleton and the microtubule network. Two methods are presented. The first, Label2Label (L2L), is a new restoration method in fluorescence microscopy that enhances the image contrast of cellular structure. L2L does not require clean target data for the training. Instead, clear differences in image quality observable between non-identical fluorescent labels are capitalised on to systematically train a network to increase structural contrast. The idea to dual-label a cell to acquire training data makes L2L relatively straightforward to implement in practice.
Moreover, results show that a multi-scale structural similarity loss function can enhance the performance in L2L when implemented carefully. Furthermore, the use of artificial labelling in interference reflection microscopy (IRM) is explored. In artificial labelling, an ANN is trained to translate between images acquired with the label-free IRM technique and fluorescence images. Both images differ in their specificity; contrast in label-free images stems from all cell components, whereas fluorescence images are ideally highly specific to the cellular target that was labelled. Consequently, with artificial labelling, specific cellular structures can be selectively visualised in label-free images. For the translation between IRM images and fluorescence images of a focal adhesion-marker, different cases were explored where target data exhibited either a low or a high contrast of the focal adhesions, and were paired or unpaired to the IRM inputs. Results presented in this work show that a network can be successfully trained for artificial labelling in IRM. Moreover, high quality paired target images are not necessarily required for that task. Here, a newly developed framework, the so-called 2LGAN, that is a variant of a generative adversarial network, performed highest for such a case. Notably, L2L and artificial labelling in IRM out-perform classical methods as pre-processing step for downstream quantitative image analysis. Both methods could have important impact in the future for the visualisation of other cellular structures not explored in this work. One such structure could be the plasma membrane which is difficult to visualise with microscopy but is of high interest in cell biology
I, victim : the creation of the ‘victim’ in Scots criminal law
Full text linkThis thesis sets out to answer the question: what is the victim as a legal construct in Scots law? This thesis demonstrates that the victim as a legal concept is protean and has changed over time, shaped and reshaped by external factors, to meet the needs of the legal system and times into which it is introduced. This thesis proposes that a new concept of the victim has been created in Scots law and will examine the socio-political factors which gave rise to this new legal phenomenon. It will then examine how international legal mechanisms enabled this new concept of the victim to permeate the centre of the Scottish criminal justice system. Finally, it will propose how this new legal construct of the victim might be understood as compatible with the framework of the current legal system, and without having to abandon the fundamental pillars of that system or creating a paradox within the system itself. Through exploring its central question, this thesis addresses three key themes: First, the relationship between the victim and the state, and specifically the impact the changing relationship between these two parties has on our understanding of the victim and the criminal justice system. Second, the impact of victim creation on our criminal justice discourse, specifically how differing attitudes of the legal function of the victim, can lead to incoherent discourse and conceptual paradox within the system itself. Thirdly, the role of a rights-based model in conceptualising the victim in Scots law, specifically the rights of recognition, access and participation and how this might offer the solution to a unifying construct of the victim.This thesis sets out to answer the question: what is the victim as a legal construct in Scots law? This thesis demonstrates that the victim as a legal concept is protean and has changed over time, shaped and reshaped by external factors, to meet the needs of the legal system and times into which it is introduced. This thesis proposes that a new concept of the victim has been created in Scots law and will examine the socio-political factors which gave rise to this new legal phenomenon. It will then examine how international legal mechanisms enabled this new concept of the victim to permeate the centre of the Scottish criminal justice system. Finally, it will propose how this new legal construct of the victim might be understood as compatible with the framework of the current legal system, and without having to abandon the fundamental pillars of that system or creating a paradox within the system itself. Through exploring its central question, this thesis addresses three key themes: First, the relationship between the victim and the state, and specifically the impact the changing relationship between these two parties has on our understanding of the victim and the criminal justice system. Second, the impact of victim creation on our criminal justice discourse, specifically how differing attitudes of the legal function of the victim, can lead to incoherent discourse and conceptual paradox within the system itself. Thirdly, the role of a rights-based model in conceptualising the victim in Scots law, specifically the rights of recognition, access and participation and how this might offer the solution to a unifying construct of the victim
How do social, economic and health policymakers consider systematic variations in population health?
Full text linkSystematic and unfair variations in population health have increased in the UK over the last decade, due to factors such as economic austerity, Covid-19 and inflation. Policy approaches to reduce such health inequalities have long faced significant challenges, including the tendency for policy to drift towards individual-level solutions, and difficulties collaborating for ‘Health in All Policies’. Research tends to focus on health policy approaches, but inequalities in health are largely shaped by inequalities in social and economic determinants. Therefore, this thesis explores social, economic and health policy approaches to health inequality. It uses two methods: firstly, a frame analysis (a deductive analysis of text according to five categories constituting a ‘policy frame’) of thirty policy strategy documents; then, semi-structured interviews with thirty-three policy actors. Texts and participants were selected from two devolved sub-state polities: Greater Manchester Combined Authority (GMCA) and the Scottish Government (SG). Documentary analysis and interviews found that health inequalities were framed instrumentally in policy texts by GMCA as part of political dialogue with national government concerning devolution. Further, three ‘health inequality’ policy frames were identified as existing across the devolved policy settings; two of which were achieving high levels of political prioritisation. Social and economic policy actors often used ‘health’ as shorthand for illness, or for health policy. This latter use surfaced tensions between policy teams that were likely to inhibit collaboration. In contrast, the term ‘wellbeing’ was widely supported as aligning closer to the social model of health than ‘health’, and because it was unaffiliated with any specific policy team. These findings imply a need for further research on how and where the term ‘health’ may have counter-productive policy impacts; and whether alternative conceptualisations may facilitate more effective policy approaches to population health.Systematic and unfair variations in population health have increased in the UK over the last decade, due to factors such as economic austerity, Covid-19 and inflation. Policy approaches to reduce such health inequalities have long faced significant challenges, including the tendency for policy to drift towards individual-level solutions, and difficulties collaborating for ‘Health in All Policies’. Research tends to focus on health policy approaches, but inequalities in health are largely shaped by inequalities in social and economic determinants. Therefore, this thesis explores social, economic and health policy approaches to health inequality. It uses two methods: firstly, a frame analysis (a deductive analysis of text according to five categories constituting a ‘policy frame’) of thirty policy strategy documents; then, semi-structured interviews with thirty-three policy actors. Texts and participants were selected from two devolved sub-state polities: Greater Manchester Combined Authority (GMCA) and the Scottish Government (SG). Documentary analysis and interviews found that health inequalities were framed instrumentally in policy texts by GMCA as part of political dialogue with national government concerning devolution. Further, three ‘health inequality’ policy frames were identified as existing across the devolved policy settings; two of which were achieving high levels of political prioritisation. Social and economic policy actors often used ‘health’ as shorthand for illness, or for health policy. This latter use surfaced tensions between policy teams that were likely to inhibit collaboration. In contrast, the term ‘wellbeing’ was widely supported as aligning closer to the social model of health than ‘health’, and because it was unaffiliated with any specific policy team. These findings imply a need for further research on how and where the term ‘health’ may have counter-productive policy impacts; and whether alternative conceptualisations may facilitate more effective policy approaches to population health
Group based physical activity interventions for people with fibromyalgia syndrome : scoping the evidence and exploring the perspectives of people living with the condition
No full textThis Master of Research thesis aimed to explore group-based physical activity (PA) interventions for people with Fibromyalgia Syndrome (FMS). Currently, there are multiple systematic reviews and meta-analyses exploring the effects of PA interventions in people with FMS. These demonstrate multiple health benefits as a direct result of increased PA, however, there are no systematic or scoping reviews explicitly investigating the effect of group-based PA interventions in people with FMS. Group-based PA interventions may prove to be a cost-effective approach to achieve similar health benefits, with the potential of ameliorating some of the barriers to engaging in PA often identified by those with FMS. Therefore, the purpose of paper one was to identify the categories of group-based PA interventions that have been investigated within the current literature and to ascertain if there were any evidence gaps which would benefit from further investigation. The key findings from this review were that seventeen studies met the inclusion criteria, four of which were RCTs, and the majority investigated multi-component interventions. This aligns with national guidelines from the National Institute for Health and Care Excellence on the management of FMS. The review found that group-based PA interventions yielded relatively low attrition rates (mean=21%), although half reported attrition greater than the generally accepted 20%, therefore it remains unclear if group-based approaches are superior to other forms of PA in reducing attrition. There is a need for further investigation into the factors which influence attrition. An exploration into the lived experience of group-based PA may be helpful to identify the important factors influencing engagement and adherence, and this may contribute to determining whether the group approach is has the potential to be more beneficial than individually delivered interventions. Therefore, paper two presents a qualitative study exploring the perspectives of adults with lived experience of FMS and their views on group-based PA interventions. Key findings from this paper suggest that people with FMS perceive there is a lack of understanding of the impact of the condition at a societal level. People with FMS have a desire to feel validated and understood. They outlined the significant challenges they face when engaging in PA and expressed a strong desire for group-based interventions that are delivered by practitioners who understand the impact of FMS and its variable and fluctuating nature. Practitioners who are empathetic, non-judgmental and foster a collaborative, open and supportive ethos which takes account of individual differences in ability and symptom experience. These factors may play a crucial role in achieving higher levels of engagement and adherence in future PA interventions.This Master of Research thesis aimed to explore group-based physical activity (PA) interventions for people with Fibromyalgia Syndrome (FMS). Currently, there are multiple systematic reviews and meta-analyses exploring the effects of PA interventions in people with FMS. These demonstrate multiple health benefits as a direct result of increased PA, however, there are no systematic or scoping reviews explicitly investigating the effect of group-based PA interventions in people with FMS. Group-based PA interventions may prove to be a cost-effective approach to achieve similar health benefits, with the potential of ameliorating some of the barriers to engaging in PA often identified by those with FMS. Therefore, the purpose of paper one was to identify the categories of group-based PA interventions that have been investigated within the current literature and to ascertain if there were any evidence gaps which would benefit from further investigation. The key findings from this review were that seventeen studies met the inclusion criteria, four of which were RCTs, and the majority investigated multi-component interventions. This aligns with national guidelines from the National Institute for Health and Care Excellence on the management of FMS. The review found that group-based PA interventions yielded relatively low attrition rates (mean=21%), although half reported attrition greater than the generally accepted 20%, therefore it remains unclear if group-based approaches are superior to other forms of PA in reducing attrition. There is a need for further investigation into the factors which influence attrition. An exploration into the lived experience of group-based PA may be helpful to identify the important factors influencing engagement and adherence, and this may contribute to determining whether the group approach is has the potential to be more beneficial than individually delivered interventions. Therefore, paper two presents a qualitative study exploring the perspectives of adults with lived experience of FMS and their views on group-based PA interventions. Key findings from this paper suggest that people with FMS perceive there is a lack of understanding of the impact of the condition at a societal level. People with FMS have a desire to feel validated and understood. They outlined the significant challenges they face when engaging in PA and expressed a strong desire for group-based interventions that are delivered by practitioners who understand the impact of FMS and its variable and fluctuating nature. Practitioners who are empathetic, non-judgmental and foster a collaborative, open and supportive ethos which takes account of individual differences in ability and symptom experience. These factors may play a crucial role in achieving higher levels of engagement and adherence in future PA interventions
Decipher, disarm and disengage : understanding the biosynthesis and self-resistance mechanisms of kirromycin-like, elfamycin producing, streptomyces.
Full text linkElfamycin antibiotics exhibit activity against Gram-positive bacteria by inhibiting translation via elongation factor EF-Tu. Elfamycins are characterised by mode of action rather than chemical structure, though minor structural modifications in these antibiotics can significantly alter their biological activity. Aurodox, a kirromycin-like elfamycin, was shown to inhibit the Type III Secretion System (T3SS) of Enteropathogenic (EPEC) and Enterohemorrhagic (EHEC) Escherichia coli along with effective anti-virulence treatment. This study aimed to investigate the diversity of kirromycin-like elfamycin biosynthetic gene clusters (BGCs), their evolution, resistance mechanisms, and how biosynthesis may be manipulated to create modified elfamycin variants which many be active against the T3SS of EHEC. Genome sequencing of the kirromycin producer Streptomyces ramocissimus was performed, and the BGC responsible for kirromycin production through comparison to the BGC found in Streptomyces collinus. Comparative analyses between kirromycin-like BGCs revealed the potential for natural variation in producing identical compounds via different genetic pathways. Additionally, a kirromycin-like BGC was identified in Streptomyces ISL094, which contained an additional methyltransferase on the BGC, similar to the aurodox BGC of S. goldiniensis, suggesting it could be an aurodox producer. Moreover, aurodox and kirromycin were found to downregulate the T3SS of EHEC, a trait previously thought to be unique to aurodox. Key genes within the aurodox BGC were manipulated and the BGC heterologously expressed to infer function of their roles in aurodox biosynthesis, where their derivatives showed similar effects on the T3SS. The protein, AurM*, thought to methylate the kirromycin molecule creating aurodox was assayed for methyltransferase activity in vitro, but activity was not demonstrated on kirromycin as a substrate. Finally, a potential "moonlighting" role for EF-Tu proteins is suggested, indicating that they might have additional functions related to T3SS regulation, where the knockdown phenotype of aurodox on the T3SS of EHEC was reversed when EF-Tu was expressed in EHEC.Elfamycin antibiotics exhibit activity against Gram-positive bacteria by inhibiting translation via elongation factor EF-Tu. Elfamycins are characterised by mode of action rather than chemical structure, though minor structural modifications in these antibiotics can significantly alter their biological activity. Aurodox, a kirromycin-like elfamycin, was shown to inhibit the Type III Secretion System (T3SS) of Enteropathogenic (EPEC) and Enterohemorrhagic (EHEC) Escherichia coli along with effective anti-virulence treatment. This study aimed to investigate the diversity of kirromycin-like elfamycin biosynthetic gene clusters (BGCs), their evolution, resistance mechanisms, and how biosynthesis may be manipulated to create modified elfamycin variants which many be active against the T3SS of EHEC. Genome sequencing of the kirromycin producer Streptomyces ramocissimus was performed, and the BGC responsible for kirromycin production through comparison to the BGC found in Streptomyces collinus. Comparative analyses between kirromycin-like BGCs revealed the potential for natural variation in producing identical compounds via different genetic pathways. Additionally, a kirromycin-like BGC was identified in Streptomyces ISL094, which contained an additional methyltransferase on the BGC, similar to the aurodox BGC of S. goldiniensis, suggesting it could be an aurodox producer. Moreover, aurodox and kirromycin were found to downregulate the T3SS of EHEC, a trait previously thought to be unique to aurodox. Key genes within the aurodox BGC were manipulated and the BGC heterologously expressed to infer function of their roles in aurodox biosynthesis, where their derivatives showed similar effects on the T3SS. The protein, AurM*, thought to methylate the kirromycin molecule creating aurodox was assayed for methyltransferase activity in vitro, but activity was not demonstrated on kirromycin as a substrate. Finally, a potential "moonlighting" role for EF-Tu proteins is suggested, indicating that they might have additional functions related to T3SS regulation, where the knockdown phenotype of aurodox on the T3SS of EHEC was reversed when EF-Tu was expressed in EHEC
Advances in the Euler-Maruyama method for stochastic differential equations with local Lipschitz coefficients
Full text linkMy PhD research is devoted to enriching the strong convergence theory of modified Euler-Maruyama methods for stochastic differential equation with locally Lipschitz coefficients. In this PhD thesis, we will introduce several modified Euler-Maruyama methods and establish their strong convergence theory. First, we will use new numerical analysis techniques to improve strong convergence results of the truncated Euler-Maruyama method. We then combine analysis techniques for polynomially growing coefficients and concave coefficients to extend the truncated EM method for multidimensional SDEs with polynomially growing drift and concave diffusion coefficients satisfying the Osgood condition. Then we will pay attention to scalar SDEs with locally Lipschitz coefficients. We will start with improving strong convergence results of the logarithmic truncated Euler-Maruyama method. To be concrete, we will use new numerical analysis techniques and further extend them for the constant elasticity of variance model and the A¨ıt-Sahalia model with almost full parameter ranges. We will prove that the logarithmic truncated Euler-Maruyama method is strongly convergent with order one half in general Lp-norm. In the rest of this thesis, we will focus on the projected Euler-Maruyama method. It has good convergence properties for scalar SDEs with locally Lipschitz coefficients. For example, it is strong Lp-convergent with order one half for the Cox-Ingersoll-Ross model with a wide parameter ranges. In particular, we will introduce a novel numerical analysis technique to prove that the projected Euler-Maruyama method may have finite inverse moments, which other modified Euler-Maruyama methods generally do not have. We will use finite inverse moments to prove that the projected Euler-Maruyama method is strong Lp - convergent with order one for many useful scalar SDE models, e.g., the constant elasticity of variance model, the A¨ıt-Sahalia model, the Heston-3/2 volatility model, the Wright-Fisher model and so on.My PhD research is devoted to enriching the strong convergence theory of modified Euler-Maruyama methods for stochastic differential equation with locally Lipschitz coefficients. In this PhD thesis, we will introduce several modified Euler-Maruyama methods and establish their strong convergence theory. First, we will use new numerical analysis techniques to improve strong convergence results of the truncated Euler-Maruyama method. We then combine analysis techniques for polynomially growing coefficients and concave coefficients to extend the truncated EM method for multidimensional SDEs with polynomially growing drift and concave diffusion coefficients satisfying the Osgood condition. Then we will pay attention to scalar SDEs with locally Lipschitz coefficients. We will start with improving strong convergence results of the logarithmic truncated Euler-Maruyama method. To be concrete, we will use new numerical analysis techniques and further extend them for the constant elasticity of variance model and the A¨ıt-Sahalia model with almost full parameter ranges. We will prove that the logarithmic truncated Euler-Maruyama method is strongly convergent with order one half in general Lp-norm. In the rest of this thesis, we will focus on the projected Euler-Maruyama method. It has good convergence properties for scalar SDEs with locally Lipschitz coefficients. For example, it is strong Lp-convergent with order one half for the Cox-Ingersoll-Ross model with a wide parameter ranges. In particular, we will introduce a novel numerical analysis technique to prove that the projected Euler-Maruyama method may have finite inverse moments, which other modified Euler-Maruyama methods generally do not have. We will use finite inverse moments to prove that the projected Euler-Maruyama method is strong Lp - convergent with order one for many useful scalar SDE models, e.g., the constant elasticity of variance model, the A¨ıt-Sahalia model, the Heston-3/2 volatility model, the Wright-Fisher model and so on
Advanced model predictive control for three-dimensional motion control of autonomous underwater vehicles
Full text linkThe increasing demand for ocean-related activities, driven by needs such as environmental preservation, offshore renewable energy deployment, border security and
weather forecasting, highlights the importance of underwater operations. With minimal human intervention, autonomous underwater vehicles (AUVs) are increasingly
employed to execute missions in water bodies. Improved AUV motion reliability
requires advanced controllers to cope with challenges posed by nonlinear dynamics,
coupled motion, actuator limits and environmental disturbances. This thesis aims to foster the use of Model Predictive Control (MPC) for AUV motion control, leveraging its capability to optimise the performance of both linear and nonlinear systems while accounting for system and operational constraints. Standard MPC uses the receding horizon strategy to offer inherent robustness under minor uncertainties. However, the effectiveness of AUV motion control in the marine environment can degrade under substantial ocean currents and wave disturbances. Moreover, the full-order nonlinear AUV model is complex, rendering it less appealing for MPC design due to the associated online computational cost. As a result, this thesis proposes formulating the full-order nonlinear AUV model as a linear parameter-varying (LPV) system. This makes obtaining a convex optimisation control problem possible, which can be efficiently solved using off-the-shelf solvers. Building on the overall research goals discussed in Chapter 1, this thesis introduces the mathematical model of an AUV in Chapter 2 and highlights issues
impacting its use in motion control design. Chapter 3 provides a state-of-the-art review of advanced predictive control methods to underscore the significance of this work. This thesis proposes three main design approaches, leveraging the LPV model, to address the effects of disturbances across various motion control tasks. The first approach resulted in two novel MPC algorithms introduced in Chapter 4, both based on velocity models that use increment variables to counteract the effects of disturbances. The first controller, LPVMPC1, is designed for dynamic positioning, while the second controller, LPVMPC2, is developed for combined dynamic positioning and trajectory tracking control of AUVs. The LPVMPC2 integrates a planning scheme to facilitate a seamless transition from the trajectory tracking task to dynamic positioning. In the LPVMPC2 design, persistent AUV operation is ensured by maintaining continuous functionality even when reference signals include unreachable positions that violate the AUV workspace constraints. The second approach, presented in Chapter 5, utilises the tube-based method for a robust tube-based MPC (TMPC) design to achieve resilience against environmental disturbances. The TMPC employs a line-of-sight (LOS) local trajectory replanning strategy to mitigate input saturation effects, enabling the consideration of realistic magnitude and rate constraints on input signals. An optimal state dependent feedback controller is proposed to construct time-varying tubes to ensure the perturbed AUV system remains within a tube centred around the nominal trajectory. The TMPC framework is computationally tractable as it requires the online
solution of a convex quadratically constrained quadratic problem. The third MPC approach is presented in Chapter 6, which introduces an enclosure based LOS guidance system and a robust min-max MPC (MM-MPC) for AUV path-following. By using the vehicle’s desired heading angle to generate reference linear and angular position coordinates, the need to formulate an AUV error model is bypassed. The simplicity of the LOS guidance system is then leveraged to develop a multi-objective LOS guidance system (MO-LOSGS) to ensure collision-free navigation amidst static obstacles. The MM-MPC is designed to stabilise the AUV speed for time- and energy-efficient navigation. The high computational cost that had limited the application of MM-MPC is mitigated by developing a duality-based transformation strategy to reformulate the problem into a quadratic minimisation control problem. All simulation validations of the developed controllers are performed using a realistic Naminow-D AUV manufactured by Mitsubishi Heavy Industries Ltd. The concluding chapter offers a summary of key research contributions to the development of advanced MPC techniques for AUV motion control and proposes potential avenues for future research. Key Words: Model Predictive Control; Dynamic Modelling; Autonomous Underwater Vehicles; Dynamic Positioning; Trajectory Tracking; Path-Following; Robust Control; Convex Optimisation.The increasing demand for ocean-related activities, driven by needs such as environmental preservation, offshore renewable energy deployment, border security and
weather forecasting, highlights the importance of underwater operations. With minimal human intervention, autonomous underwater vehicles (AUVs) are increasingly
employed to execute missions in water bodies. Improved AUV motion reliability
requires advanced controllers to cope with challenges posed by nonlinear dynamics,
coupled motion, actuator limits and environmental disturbances. This thesis aims to foster the use of Model Predictive Control (MPC) for AUV motion control, leveraging its capability to optimise the performance of both linear and nonlinear systems while accounting for system and operational constraints. Standard MPC uses the receding horizon strategy to offer inherent robustness under minor uncertainties. However, the effectiveness of AUV motion control in the marine environment can degrade under substantial ocean currents and wave disturbances. Moreover, the full-order nonlinear AUV model is complex, rendering it less appealing for MPC design due to the associated online computational cost. As a result, this thesis proposes formulating the full-order nonlinear AUV model as a linear parameter-varying (LPV) system. This makes obtaining a convex optimisation control problem possible, which can be efficiently solved using off-the-shelf solvers. Building on the overall research goals discussed in Chapter 1, this thesis introduces the mathematical model of an AUV in Chapter 2 and highlights issues
impacting its use in motion control design. Chapter 3 provides a state-of-the-art review of advanced predictive control methods to underscore the significance of this work. This thesis proposes three main design approaches, leveraging the LPV model, to address the effects of disturbances across various motion control tasks. The first approach resulted in two novel MPC algorithms introduced in Chapter 4, both based on velocity models that use increment variables to counteract the effects of disturbances. The first controller, LPVMPC1, is designed for dynamic positioning, while the second controller, LPVMPC2, is developed for combined dynamic positioning and trajectory tracking control of AUVs. The LPVMPC2 integrates a planning scheme to facilitate a seamless transition from the trajectory tracking task to dynamic positioning. In the LPVMPC2 design, persistent AUV operation is ensured by maintaining continuous functionality even when reference signals include unreachable positions that violate the AUV workspace constraints. The second approach, presented in Chapter 5, utilises the tube-based method for a robust tube-based MPC (TMPC) design to achieve resilience against environmental disturbances. The TMPC employs a line-of-sight (LOS) local trajectory replanning strategy to mitigate input saturation effects, enabling the consideration of realistic magnitude and rate constraints on input signals. An optimal state dependent feedback controller is proposed to construct time-varying tubes to ensure the perturbed AUV system remains within a tube centred around the nominal trajectory. The TMPC framework is computationally tractable as it requires the online
solution of a convex quadratically constrained quadratic problem. The third MPC approach is presented in Chapter 6, which introduces an enclosure based LOS guidance system and a robust min-max MPC (MM-MPC) for AUV path-following. By using the vehicle’s desired heading angle to generate reference linear and angular position coordinates, the need to formulate an AUV error model is bypassed. The simplicity of the LOS guidance system is then leveraged to develop a multi-objective LOS guidance system (MO-LOSGS) to ensure collision-free navigation amidst static obstacles. The MM-MPC is designed to stabilise the AUV speed for time- and energy-efficient navigation. The high computational cost that had limited the application of MM-MPC is mitigated by developing a duality-based transformation strategy to reformulate the problem into a quadratic minimisation control problem. All simulation validations of the developed controllers are performed using a realistic Naminow-D AUV manufactured by Mitsubishi Heavy Industries Ltd. The concluding chapter offers a summary of key research contributions to the development of advanced MPC techniques for AUV motion control and proposes potential avenues for future research. Key Words: Model Predictive Control; Dynamic Modelling; Autonomous Underwater Vehicles; Dynamic Positioning; Trajectory Tracking; Path-Following; Robust Control; Convex Optimisation
Comparing factor models in European stock returns
Full text linkThe goal of asset pricing research is to find the optimal model that explains the drivers of asset
returns. Historically, this field has predominantly relied on data from the United States, given
the extensive and detailed records of its financial markets. Due to the growing interdependence
of international markets, recent research has shifted towards leveraging large global datasets to
develop universally applicable models. However, empirical evidence suggests that these global
models explain less variation in domestic returns compared to country-specific models.
This thesis investigates the effectiveness of country-specific asset pricing models across a set
of European markets, utilising both classical and Bayesian methods to assess model
performance. The first empirical chapter begins with evaluating the relative performance of
nine asset pricing models in developed European stock markets from 1991-2022.
Asymptotically valid tests of model comparison, developed by Barillas, Kan, Robotti and
Shanken (2020), are conducted, where the extent of model mispricing is gauged by the squared
Sharpe ratio improvement measure of Barillas and Shanken (2017). The findings reveal that
the Fama and French (2018) six-factor model, with both original and updated value factors, are
the top-performing models in most markets. However, variation in the absolute and relative
performance of models across samples suggests that a singular optimal European asset pricing
model does not exist within the classical framework.
To enhance model performance, the second empirical chapter explores the use of serial
correlation in factor returns as conditioning information. Adopting the methodology of Ehsani
and Linnainmaa (2022), this chapter shows that multiple investment factors in the crosscountry dataset are unconditionally minimum-variance inefficient: factor returns are positively
autocorrelated, while risk remains constant regardless of past returns. Using Ferson and
Siegel’s (2001) general framework, 'time-series efficient factors' are constructed by
conditioning factor weights on historical returns to enhance the Sharpe ratios of these factors
across the European markets under consideration. A number of these optimised factors achieve
significantly higher average Sharpe ratios compared to the original factors, while retaining all
the information contained in the original factors. When the model comparison tests of Barillas
et al. (2020) are repeated with these optimised factors, the absolute performance of the lowerperforming models improves, while the relative performance among the models remains
consistent across markets.
In the third and final empirical chapter, the Bayesian framework of Chib, Zeng, and Zhao
(2020) is used to identify the optimal combination of factors from a starting collection of 12
risk factors in each European market. The results indicate that the optimal combinations of
factors are similar to the top-performing models in the classical tests. The optimal model from
the scan either represents a reduced form with one or two fewer factors or an extension of the
top model identified in Chapter Two, with one or two additional factors. This alignment
underscores the robustness of the model selection across different testing methodologies. The
changes in these optimal combinations are then examined under the assumptions of both
normality and multivariate-t distributions on the factor data. Employing the methodology of
Chib and Zeng (2020), the analysis reveals no significant disparities in results when a Studentt distribution is assumed for the factor data. Additionally, the extent to which the efficient factor
transformation impacts the model comparison tests in each market is analysed. The findings
reveal that certain efficient factors are present in the optimal combination of factors across
European markets.The goal of asset pricing research is to find the optimal model that explains the drivers of asset
returns. Historically, this field has predominantly relied on data from the United States, given
the extensive and detailed records of its financial markets. Due to the growing interdependence
of international markets, recent research has shifted towards leveraging large global datasets to
develop universally applicable models. However, empirical evidence suggests that these global
models explain less variation in domestic returns compared to country-specific models.
This thesis investigates the effectiveness of country-specific asset pricing models across a set
of European markets, utilising both classical and Bayesian methods to assess model
performance. The first empirical chapter begins with evaluating the relative performance of
nine asset pricing models in developed European stock markets from 1991-2022.
Asymptotically valid tests of model comparison, developed by Barillas, Kan, Robotti and
Shanken (2020), are conducted, where the extent of model mispricing is gauged by the squared
Sharpe ratio improvement measure of Barillas and Shanken (2017). The findings reveal that
the Fama and French (2018) six-factor model, with both original and updated value factors, are
the top-performing models in most markets. However, variation in the absolute and relative
performance of models across samples suggests that a singular optimal European asset pricing
model does not exist within the classical framework.
To enhance model performance, the second empirical chapter explores the use of serial
correlation in factor returns as conditioning information. Adopting the methodology of Ehsani
and Linnainmaa (2022), this chapter shows that multiple investment factors in the crosscountry dataset are unconditionally minimum-variance inefficient: factor returns are positively
autocorrelated, while risk remains constant regardless of past returns. Using Ferson and
Siegel’s (2001) general framework, 'time-series efficient factors' are constructed by
conditioning factor weights on historical returns to enhance the Sharpe ratios of these factors
across the European markets under consideration. A number of these optimised factors achieve
significantly higher average Sharpe ratios compared to the original factors, while retaining all
the information contained in the original factors. When the model comparison tests of Barillas
et al. (2020) are repeated with these optimised factors, the absolute performance of the lowerperforming models improves, while the relative performance among the models remains
consistent across markets.
In the third and final empirical chapter, the Bayesian framework of Chib, Zeng, and Zhao
(2020) is used to identify the optimal combination of factors from a starting collection of 12
risk factors in each European market. The results indicate that the optimal combinations of
factors are similar to the top-performing models in the classical tests. The optimal model from
the scan either represents a reduced form with one or two fewer factors or an extension of the
top model identified in Chapter Two, with one or two additional factors. This alignment
underscores the robustness of the model selection across different testing methodologies. The
changes in these optimal combinations are then examined under the assumptions of both
normality and multivariate-t distributions on the factor data. Employing the methodology of
Chib and Zeng (2020), the analysis reveals no significant disparities in results when a Studentt distribution is assumed for the factor data. Additionally, the extent to which the efficient factor
transformation impacts the model comparison tests in each market is analysed. The findings
reveal that certain efficient factors are present in the optimal combination of factors across
European markets
The role of first-line managers in implementing direct employee voice processes formally and informally in the banking sector of the Sultanate of Oman
Full text linkThis study investigates the role of first-line managers (FLMs) in implementing direct
employee voice in the private sector, focusing on how FLMs encourage and manage
both formal and informal forms of employee participation. Direct voice is defined as
a two-way communication between management and individual employees without
the mediation of a third party, such as unions or collective bargaining processes.
The study is situated in a private sector organisation, where employee voice is
progressively recognised as crucial for improving workplace efficiency, productivity,
and general involvement. It examines the interactions between FLMs, human
resources (HR) managers, and employees to shed light on how employees are
encouraged to express ideas, suggestions, and concerns, and how these
contributions affect individual and organisational performance.
The study’s key findings reflect a variety of organisational dynamics. While some
FLMs successfully implement HR policies regarding employee voice, others struggle
due to lacking resources, training, and managerial skills. A noticeable finding is that
employee participation and suggestion result in enhanced engagement, satisfaction,
and productivity. However, variations in managerial approaches to employee
suggestions can have a negative impact on employee attitudes and satisfaction. The
study also recognises challenges among FLMs, such as role ambiguity and
insufficient people management skills, which impede effective HR implementation.
This study improves the understanding of employee voice by applying ability,
motivation, and opportunity theory to the roles of FLMs and their employees. It
demonstrates that, for FLMs to effectively manage employee voice, they must have
the necessary abilities and motivation, and the right opportunities to facilitate it. The findings add to the expanding literature on employee voice by emphasising FLMs’
vital role as key facilitators of employee voice. Furthermore, this study finds that,
while FLMs are supported by their managers or department heads, they frequently
lack support from HR, highlighting critical areas for improvement in HR practices and
training.
Additionally, this thesis adopted a qualitative single case study design to investigate
how FLMs implement direct employee voice processes. Exploring multiple
viewpoints allows the researcher to grasp a greater understanding of the subject,
and therefore qualitative data was gathering from all three levels (HR managers,
FLMs, and employees) from different occupational groups at one of the leading
banks in Oman. In this study, manual thematic analysis is adopted to analyse the
responses presented by interviewees. Moreover, a combined discussion of the
empirical findings of all parties is utilised to obtain an overview of the emerging
themes from the interview questions at the three levels described.
Furthermore, the theoretical background associated with qualitative data analyses
and the adoption of the manual thematic analysis is explained. Following this, the
process of generating the initial themes is detailed. The findings derived from the
open-ended questions of the semi-structured interviews are reported. Finally, in the
conclusion of this study, the contributions to academic knowledge, suggested
actions for organisations to pursue to better facilitate their business model and
practice, and ideas for future research are provided.This study investigates the role of first-line managers (FLMs) in implementing direct
employee voice in the private sector, focusing on how FLMs encourage and manage
both formal and informal forms of employee participation. Direct voice is defined as
a two-way communication between management and individual employees without
the mediation of a third party, such as unions or collective bargaining processes.
The study is situated in a private sector organisation, where employee voice is
progressively recognised as crucial for improving workplace efficiency, productivity,
and general involvement. It examines the interactions between FLMs, human
resources (HR) managers, and employees to shed light on how employees are
encouraged to express ideas, suggestions, and concerns, and how these
contributions affect individual and organisational performance.
The study’s key findings reflect a variety of organisational dynamics. While some
FLMs successfully implement HR policies regarding employee voice, others struggle
due to lacking resources, training, and managerial skills. A noticeable finding is that
employee participation and suggestion result in enhanced engagement, satisfaction,
and productivity. However, variations in managerial approaches to employee
suggestions can have a negative impact on employee attitudes and satisfaction. The
study also recognises challenges among FLMs, such as role ambiguity and
insufficient people management skills, which impede effective HR implementation.
This study improves the understanding of employee voice by applying ability,
motivation, and opportunity theory to the roles of FLMs and their employees. It
demonstrates that, for FLMs to effectively manage employee voice, they must have
the necessary abilities and motivation, and the right opportunities to facilitate it. The findings add to the expanding literature on employee voice by emphasising FLMs’
vital role as key facilitators of employee voice. Furthermore, this study finds that,
while FLMs are supported by their managers or department heads, they frequently
lack support from HR, highlighting critical areas for improvement in HR practices and
training.
Additionally, this thesis adopted a qualitative single case study design to investigate
how FLMs implement direct employee voice processes. Exploring multiple
viewpoints allows the researcher to grasp a greater understanding of the subject,
and therefore qualitative data was gathering from all three levels (HR managers,
FLMs, and employees) from different occupational groups at one of the leading
banks in Oman. In this study, manual thematic analysis is adopted to analyse the
responses presented by interviewees. Moreover, a combined discussion of the
empirical findings of all parties is utilised to obtain an overview of the emerging
themes from the interview questions at the three levels described.
Furthermore, the theoretical background associated with qualitative data analyses
and the adoption of the manual thematic analysis is explained. Following this, the
process of generating the initial themes is detailed. The findings derived from the
open-ended questions of the semi-structured interviews are reported. Finally, in the
conclusion of this study, the contributions to academic knowledge, suggested
actions for organisations to pursue to better facilitate their business model and
practice, and ideas for future research are provided
Access Repository Dashboard
Do you manage STAX (Strathclyde Repository)? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!