157082 research outputs found
Sort by
Quantifiers for Differentiable Logics in Rocq
No full textThe interpretation of logical expressions into loss functions has given rise to so-called differentiable logics. They function as a bridge between formal logic and machine learning, offering a novel approach for property-driven training. The added expressiveness of these logics comes at the price of a more intricate semantics for first-order quantifiers. To ease their integration into machine-learning backends, we explore how to formalize semantics for first-order differentiable logics using the Mathematical Components library in the Rocq proof assistant. We seek to give rigorous semantics for quantifiers, verify their properties with respect to other logical connectives, as well as prove the soundness and completeness of the resulting logics.</p
Towards reliable subsea object recovery: a simulation study of an auv with a suction-actuated end effector
No full textAutonomous object recovery in the hadal zone is challenging due to extreme hydrostatic pressure, limited visibility and currents, and the need for precise manipulation at full ocean depth. Field experimentation in such environments is costly, high-risk, and constrained by limited vehicle availability, making early validation of autonomous behaviors difficult. This paper presents a simulation-based study of a complete autonomous subsea object recovery mission using a Hadal Small Vehicle (HSV) equipped with a three-degree-of-freedom robotic arm and a suction-actuated end effector. The Stonefish simulator is used to model realistic vehicle dynamics, hydrodynamic disturbances, sensing, and interaction with a target object under hadal-like conditions. The control framework combines a world-frame PID controller for vehicle navigation and stabilization with an inverse-kinematics-based manipulator controller augmented by acceleration feed-forward, enabling coordinated vehicle - manipulator operation. In simulation, the HSV autonomously descends from the sea surface to 6,000 m, performs structured seafloor coverage, detects a target object, and executes a suction-based recovery. The results demonstrate that high-fidelity simulation provides an effective and low-risk means of evaluating autonomous deep-sea intervention behaviors prior to field deployment
Fish Diversity and Invasive Species Detection in the Perak River, Malaysia, Through eDNA Metabarcoding
No full textFreshwater ecosystems are among the most threatened globally, facing rapid biodiversity declines due to habitat degradation, pollution, overexploitation, and invasive species introductions. Southeast Asia, recognized as a global freshwater biodiversity hotspot, is experiencing some of the highest extinction risks. The Perak River, Malaysia’s second-longest river, supports diverse fish communities critical for regional ecological stability and food security, but comprehensive biodiversity assessments have been lacking. This study used environmental DNA (eDNA) metabarcoding targeting the 12S rRNA gene to comprehensively assess fish diversity and detect invasive species across the Perak River. Water samples from five locations were analyzed using high-throughput sequencing on the Illumina MiSeq platform. We identified 16 orders, 32 families, 51 genera, and 57 species, with Cyprinidae being the dominant family (15.78% of species). Invasive species such as Micropterus salmoides, Oreochromis spp., and Gambusia affinis recognised as invasive either globally or in Malaysia were also detected, raising ecological concerns. Our results highlight the efficiency of eDNA metabarcoding for non-invasive biodiversity monitoring in tropical river systems and demonstrate its potential for early detection of invasive species. This pioneering baseline study provides essential data to guide conservation strategies, inform management policies, and contribute to broader efforts addressing freshwater biodiversity loss in Southeast Asia
Techno-economic Evaluation of the Effects of Decarbonisation and Waste Heat Utilisation on Green Hydrogen Produced from Microalgae Steam Gasification
No full textA comprehensive techno-economic evaluation of the application of steam gasification to different microalgae for hydrogen production was performed. The production of hydrogen-rich syngas from three microalgal feedstocks (namely S. Acuminatus; S. Almeriensis and a co-culture of the two microalgae) was firstly considered and then separated to produce hydrogen. The flue gases were decarbonised using Calcium Looping Combustion (CaLC) process. Since decarbonisation comes with a significant energy penalty, waste heat utilization (WHU) was applied to the CaLC for combined heat and power generation. S. Acuminatus produced the highest amount of syngas and correspondingly the greatest amount of hydrogen (157.3 kg hr−1). The yields of the produced decarbonised hydrogen of the three microalgae ranged from 13.4 to 16.5%, higher than published values ranging from 8.0 to 12.0%. Since S. Acuminatus showed the most favourable technical performance, an economic analysis was performed, resulting in levelised costs of hydrogen (LCOH) ranging from 2.46 to 3.87 $/kg. Optimising the operation conditions of CaLC process resulted in an overall decarbonisation of 97% with final carbon emission intensities less than 0.5 kg CO2/kg H2, less than reported intensities for green hydrogen production. Implementing CaLC process resulted in energy penalties ranging from 10.36 to 11.24%. Nevertheless, applying WHU resulted in overall surplus in energy generation and overall plant efficiency. Finally, CO2 abatement costs were estimated and found to be comparable to published values. Overall, this research proves that steam gasification of microalgae combined with CaLC technology is a feasible option for green hydrogen production
Lost in the Story: The Impact of Narrative with a Direction-Giving Robot
No full textSharing a story alongside an expository response is inherently human, often enhancing communication by adding personal details based on our unique experiences to what we say. When used in a task environment, narratives may be used to exploit measurable effects, such as on memory recall or interaction engagement. With the increasing presence of social robots in everyday environments, it remains unclear whether narrative communication from robots (e.g. “This picture shows a family who recently...”) instead of a factual description yields similar benefits to those observed in human-human interactions. In this paper, we develop and study a direction-giving robot, comparing three styles of navigation instruction: narrative with landmarks, landmarks only, and baseline without landmarks. We evaluate the effects of these conditions on recall, task success, and social acceptability factors (N=38) using a Furhat robot receptionist in a lab environment.Our findings show that landmark-based navigation significantly enhances perceived usefulness, task success, and social acceptability compared to baseline. Furthermore, narrative-based navigation led to significantly higher recall of individual landmarks and improved perceptions of the robot’s adaptability. These results suggest that narratives can play a practical role in enhancing task-based HRI, particularly in scenarios that demand long-term engagement, user-centered adaptability, or memory retention (e.g. education, or explainability in AI systems).This work contributes to the broader conversation on incorporating human-like conversational features in robots and highlights narratives as a potential tool for designing more effective human-robot interactions
Temporal, spectral, and modal analysis of a thin-slab Tm:LuLF hybrid stable-unstable laser
No full textTemporal instabilities observed in a thin-slab Tm:LuLF hybrid stable-unstable resonator are studied both experimentally and with numerical modelling. Near-field beam profiles and intra-cavity modes are investigated under different alignment conditions, and a two-dimensional Fox and Li model is developed and applied to verify the results. Laser rate equations for the gain medium can be integrated into the model to understand how Thulium’s complex energy structure may be influencing the resonator dynamics. The laser was scaled to ~30 W of high-brightness output power
Explicit Non-Abelian Gerbes with Connections
No full textWe define the notion of adjustment for strict Lie 2-groups and provide the complete cocycle description for non-Abelian gerbes with connections whose structure 2-group is an adjusted 2-group. Most importantly, we depart from the common fake-flat connections and employ adjusted connections. This is an important generalisation that is needed for physical applications especially in the context of supergravity. We give a number of explicit examples; in particular, we lift the spin structure on S4, corresponding to an instanton–anti-instanton pair, to a string structure, a 2-group bundle with connection. We also outline how categorified forms of Bogomolny monopoles known as self-dual strings can be obtained via a Penrose–Ward transform of string bundles over twistor space
The role of structural compartmentalization in carbon storage site selection in Permian (Rotliegend Group) reservoirs in the Southern North Sea
No full textThe Southern North Sea (SNS) gas basin is a key area for CO2 storage projects in the UK. Many of the now-depleted Permian (Rotliegend) Leman Sandstone Formation fields are being re-evaluated as carbon stores. However, the reservoir is known to be highly faulted, often leading to field compartmentalization. This has historically impacted field development and production, and will challenge suitability for CO2 storage by limiting site capacity, requiring a high number of injector wells, and increasing capital costs. It is necessary to understand the nature of these pressure compartments - and whether any individual culmination can house sufficient capacity - before devising a carbon storage development programme. The highly compartmentalized Indefatigable (Inde) Field was evaluated as a case study. A static model of the field was constructed using 3D seismic and well log data, and subsequently used to calculate the CO2 capacity of each of the 12 compartments. Five compartments were found to have capacities larger than 10 Mt, with the large Main Horst found to host 51% of total CO2 capacity. A sequential filling-and-sealing storage site development plan is suggested based on the evaluation and ranking of these compartments
The role of structural compartmentalization in carbon storage site selection in Permian (Rotliegend Group) reservoirs in the Southern North Sea
No full textThe Southern North Sea (SNS) gas basin is a key area for CO2 storage projects in the UK. Many of the now-depleted Permian (Rotliegend) Leman Sandstone Formation fields are being re-evaluated as carbon stores. However, the reservoir is known to be highly faulted, often leading to field compartmentalization. This has historically impacted field development and production, and will challenge suitability for CO2 storage by limiting site capacity, requiring a high number of injector wells, and increasing capital costs. It is necessary to understand the nature of these pressure compartments - and whether any individual culmination can house sufficient capacity - before devising a carbon storage development programme. The highly compartmentalized Indefatigable (Inde) Field was evaluated as a case study. A static model of the field was constructed using 3D seismic and well log data, and subsequently used to calculate the CO2 capacity of each of the 12 compartments. Five compartments were found to have capacities larger than 10 Mt, with the large Main Horst found to host 51% of total CO2 capacity. A sequential filling-and-sealing storage site development plan is suggested based on the evaluation and ranking of these compartments
Accelerated and enhanced multiplicative deblurring schemes
No full textWe propose modifications of the Richardson–Lucy iterations (RL) and the Image Space Reconstruction Algorithm (ISRA) that demonstrate accelerated convergence and lead to improved image restoration results. We show that the iterations of RL, ISRA, and the proposed modifications can be interpreted as fixed-point iterations corresponding to the minimizers of certain variational problems. We demonstrate that combining each iteration of the proposed modifications with an adaptive image smoothing procedure leads to substantial improvements of the image restoration results. An implementation is available at https://github.com/fayolle/Mult_BBDeblur_demo
Access Repository Dashboard
Do you manage Heriot Watt Pure? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!