User Intervention Research Articles

Empowering digital twins with large language models for global temporal feature learning

Digital Twin (DT), as an efficient technology for virtual-physical interaction, has demonstrated significant application potential in various industries. Intelligent agent-driven digital twin systems excel in analysis, decision-making, and control, making them highly suitable for manufacturing resource scheduling, diagnostic decision-making, and other requirements. However, current intelligent agents have notable deficiencies in adaptability, data utilization, and interpretability. This limitation undermines decision security and acceptability, creating barriers for user intervention. Therefore, this paper introduces a DT multi agent architecture driven by Large Language Models (LLMs). Agents perceive the characteristics of physical systems, particularly their temporal characteristics, by integrating data from various modalities. Multiple agents achieve insights through specific interaction mechanisms, while maintaining traceability. To showcase the advantages and characteristics of this architecture, we developed a typical application scenario for equipment maintenance. The effectiveness of each framework component was validated through ablation experiments. The experimental results suggest that the proposed framework holds promising and extensive application prospects.

An intelligent assistive driving solution based on smartphone for power wheelchair mobility

This research introduces a cost-effective, smartphone-powered, computer-vision-based system for Power Wheelchair Intelligent Assistive Driving (PWC IA-Driving). This system enables the safe, hands-free operation of a Power Wheelchair (PWC) with reduced attention required from the user in indoor environments, thereby easing the burden on disabled individuals and lessening their stress. Our objective is to provide an affordable and practical solution that can be easily incorporated into existing PWCs. The system leverages a customized and pre-trained ResNet-based model on a smartphone to derive driving commands from real-time imagery captured by the phone’s camera. These instructions are then conveyed to the PWC via a control interface connected to the smartphone. We have developed a prototype of this assistive driving system on a Pixel-6 Android phone and tested its feasibility on a mobile robot as a proof-of-concept. Our evaluations indicate that the smartphone can process up to 25 images per second. This rapid processing rate allows for the generation of driving instructions in real time, enabling the mobile robot to navigate safely at reasonable speeds within the test environment, while requiring minimal user intervention.

Using Solidworks Simulation tool for automated design of drying chambers and study of their operation parameters

The paper proposes a modern approach to the automated design of three-dimensional models of wood drying chambers and the study of their operation parameters using the SolidWorks Simulation tool, which makes it possible to create various information models using computer tools. Therefore, in this paper, a three-dimensional model of a wood drying chamber was developed and presented, which includes its main components: walls, ceilings, door, stacks of lumber to be dried, a fresh air supply system, a system for its circulation through fans and heating through heaters. For the automated design of three-dimensional models of all these components, the corresponding algorithms were also developed and programmatically implemented using a specialized SolidWorks application programming interface. At the same time, the developed software application takes into account the parameters entered by the user, including the size, quantity, and type of lumber to be dried. In addition, several thermal calculations were performed across the stack, including with different air velocities. This allows us to analyze the effect of this speed on the distribution of heat transfer across the stack. In general, the obtained results are automatically displayed in one of the windows of the software application. This approach makes it possible to obtain results in the background of SolidWorks with minimal user intervention. Despite the fact that such studies are carried out in SolidWorks Simulation, the graphical results, including the temperature distribution across the stack, are fully automated without user intervention. The software application also makes it possible to automatically change the geometry of a three-dimensional model of a wood drying chamber or stacks within specified limits. In turn, the use of the developed software application in combination with SolidWorks Simulation can effectively speed up the process of creating three-dimensional models of wood drying chambers and studying their operation parameters. In general, this work reflects the trend towards digital transformation in industry, opening up new opportunities for woodworking enterprises and contributing to their market competitiveness and production efficiency.

CUQIpy: I. Computational uncertainty quantification for inverse problems in Python

This paper introduces CUQIpy, a versatile open-source Python package for computational uncertainty quantification (UQ) in inverse problems, presented as Part I of a two-part series. CUQIpy employs a Bayesian framework, integrating prior knowledge with observed data to produce posterior probability distributions that characterize the uncertainty in computed solutions to inverse problems. The package offers a high-level modeling framework with concise syntax, allowing users to easily specify their inverse problems, prior information, and statistical assumptions. CUQIpy supports a range of efficient sampling strategies and is designed to handle large-scale problems. Notably, the automatic sampler selection feature analyzes the problem structure and chooses a suitable sampler without user intervention, streamlining the process. With a selection of probability distributions, test problems, computational methods, and visualization tools, CUQIpy serves as a powerful, flexible, and adaptable tool for UQ in a wide selection of inverse problems. Part II of the series focuses on the use of CUQIpy for UQ in inverse problems with partial differential equations.

Supported decision-making interventions in mental healthcare: A systematic review of current evidence and implementation barriers.

There is a growing momentum around the world to foster greater opportunities for the involvement of mental health service users in their care and treatment planning. In-principle support for this aim is widespread across mental healthcare professionals. Yet, progress in mental health services towards this objective has lagged in practice. We conducted a systematic review of quantitative, qualitative and mixed-method research on interventions to improve opportunities for the involvement of mental healthcare service users in treatment planning, to understand the current research evidence and the barriers to implementation. Seven databases were searched and 5137 articles were screened. Articles were included if they reported on an intervention for adult service users, were published between 2008 and October 2023 and were in English. Evidence in the 140 included articles was synthesised according to the JBI guidance on Mixed Methods Systematic Reviews. Research in this field remains exploratory in nature, with a wide range of interventions investigated to date but little experimental replication. Overarching barriers to shared and supported decision-making in mental health treatment planning were (1) Organisational (resource limitations, culture barriers, risk management priorities and structure); (2) Process (lack of knowledge, time constraints, health-related concerns, problems completing and using plans); and (3) Relationship barriers (fear and distrust for both service users and clinicians). On the basis of the barriers identified, recommendations are made to enable the implementation of new policies and programs, the designing of new tools and for clinicians seeking to practice shared and supported decision-making in the healthcare they offer. This systematic review has been guided at all stages by a researcher with experience of mental health service use, who does not wish to be identified at this point in time. The findings may inform organisations, researchers and practitioners on implementing supported decision-making, for the greater involvement of people with mental ill health in their healthcare.

Restorative Justice Approach To Victimless Crime

Restorative Justice Approach To Victimless Crime

Navigating autonomy and control in human-AI delegation: User responses to technology- versus user-invoked task allocation

Users can increasingly delegate to information systems (IS) – that is transferring rights and responsibilities regarding certain tasks – even to the degree that IS can act autonomously (i.e., without the intervention or supervision of users). What is more, IS increasingly offer to assume the rights and responsibilities for a task not only in response to user prompts (i.e., user-invoked delegation) but also without user prompts (i.e., IS-invoked delegation). Yet, little is known about whether, how, and why users agree to delegation when they are asked by the IS in contrast to when they self-initiate the delegation. Drawing on self-affirmation theory, we investigate user acceptance of IS- versus user-invoked delegation in two complementary online experiments in software development. Our core findings reveal that IS-invoked (vs. user-invoked) delegation increases users' perceived self-threat and thus decreases their willingness to accept delegation. This threatening effect is larger the less (vs. more) the user perceives control after the potential delegation. Taken together, we uncover defensive user responses to IS-invoked delegation. Furthermore, we shed light on the underlying and moderating mechanisms representing the reasons and contextual features that explain and mitigate these defensive measures. These findings have significant implications for IS designers seeking to improve user-IS collaboration and outcomes by employing IS-invoked delegation.

Concept logic trees: enabling user interaction for transparent image classification and human-in-the-loop learning

Interpretable deep learning models are increasingly important in domains where transparent decision-making is required. In this field, the interaction of the user with the model can contribute to the interpretability of the model. In this research work, we present an innovative approach that combines soft decision trees, neural symbolic learning, and concept learning to create an image classification model that enhances interpretability and user interaction, control, and intervention. The key novelty of our method relies on the fusion of an interpretable architecture with neural symbolic learning, allowing the incorporation of expert knowledge and user interaction. Furthermore, our solution facilitates the inspection of the model through queries in the form of first-order logic predicates. Our main contribution is a human-in-the-loop model as a result of the fusion of neural symbolic learning and an interpretable architecture. We validate the effectiveness of our approach through comprehensive experimental results, demonstrating competitive performance on challenging datasets when compared to state-of-the-art solutions.

ColabSeg: An interactive tool for editing, processing, and visualizing membrane segmentations from cryo-ET data

Cellular cryo-electron tomography (cryo-ET) has emerged as a key method to unravel the spatial and structural complexity of cells in their near-native state at unprecedented molecular resolution. To enable quantitative analysis of the complex shapes and morphologies of lipid membranes, the noisy three-dimensional (3D) volumes must be segmented. Despite recent advances, this task often requires considerable user intervention to curate the resulting segmentations. Here, we present ColabSeg, a Python-based tool for processing, visualizing, editing, and fitting membrane segmentations from cryo-ET data for downstream analysis. ColabSeg makes many well-established algorithms for point-cloud processing easily available to the broad community of structural biologists for applications in cryo-ET through its graphical user interface (GUI). We demonstrate the usefulness of the tool with a range of use cases and biological examples. Finally, for a large Mycoplasma pneumoniae dataset of 50 tomograms, we show how ColabSeg enables high-throughput membrane segmentation, which can be used as valuable training data for fully automated convolutional neural network (CNN)-based segmentation.

NeuroEditor: a tool to edit and visualize neuronal morphologies.

The digital extraction of detailed neuronal morphologies from microscopy data is an essential step in the study of neurons. Ever since Cajal's work, the acquisition and analysis of neuron anatomy has yielded invaluable insight into the nervous system, which has led to our present understanding of many structural and functional aspects of the brain and the nervous system, well beyond the anatomical perspective. Obtaining detailed anatomical data, though, is not a simple task. Despite recent progress, acquiring neuron details still involves using labor-intensive, error prone methods that facilitate the introduction of inaccuracies and mistakes. In consequence, getting reliable morphological tracings usually needs the completion of post-processing steps that require user intervention to ensure the extracted data accuracy. Within this framework, this paper presents NeuroEditor, a new software tool for visualization, editing and correction of previously reconstructed neuronal tracings. This tool has been developed specifically for alleviating the burden associated with the acquisition of detailed morphologies. NeuroEditor offers a set of algorithms that can automatically detect the presence of potential errors in tracings. The tool facilitates users to explore an error with a simple mouse click so that it can be corrected manually or, where applicable, automatically. In some cases, this tool can also propose a set of actions to automatically correct a particular type of error. Additionally, this tool allows users to visualize and compare the original and modified tracings, also providing a 3D mesh that approximates the neuronal membrane. The approximation of this mesh is computed and recomputed on-the-fly, reflecting any instantaneous changes during the tracing process. Moreover, NeuroEditor can be easily extended by users, who can program their own algorithms in Python and run them within the tool. Last, this paper includes an example showing how users can easily define a customized workflow by applying a sequence of editing operations. The edited morphology can then be stored, together with the corresponding 3D mesh that approximates the neuronal membrane.

Automated eccentricity measurement from raw eclipsing binary light curves with intrinsic variability

Eclipsing binary systems provide the opportunity to measure the fundamental parameters of their component stars in a stellar-model-independent way. This makes them ideal candidates for testing and calibrating theories of stellar structure and (tidal) evolution. Large photometric (space) surveys provide a wealth of data for both the discovery and the analysis of these systems. Even without spectroscopic follow-up there is often enough information in their photometric time series to warrant analysis, especially if there is an added value present in the form of intrinsic variability, such as pulsations. Our goal is to implement and validate a framework for the homogeneous analysis of large numbers of eclipsing binary light curves, such as the numerous high-duty-cycle observations from space missions like TESS. The aim of this framework is to be quick and simple to run and to limit the user's time investment when obtaining, amongst other parameters, orbital eccentricities. We developed a new and fully automated methodology for the analysis of eclipsing binary light curves with or without additional intrinsic variability. Our method includes a fast iterative pre-whitening procedure that results in a list of extracted sinusoids that is broadly applicable for purposes other than eclipses. After eclipses are identified and measured, orbital and stellar parameters are measured under the assumption of spherical stars of uniform brightness. We tested our methodology in two settings: a set of synthetic light curves with known input and the catalogue of Kepler eclipsing binaries. The synthetic tests show that we can reliably recover the frequencies and amplitudes of the sinusoids included in the signal as well as the input binary parameters, albeit to varying degrees of accuracy. Recovery of the tangential component of eccentricity is the most accurate and precise. Kepler results confirm a robust determination of orbital periods, with 80.5<!PCT!> of periods matching the catalogued ones. We present the eccentricities for this analysis and show that they broadly follow the theoretically expected pattern as a function of the orbital period. Our analysis methodology is shown to be capable of analysing large numbers of eclipsing binary light curves with no user intervention, and in doing so provide a basis for a further in-depth analysis of systems of particular interest as well as for statistical analysis at the sample level. Furthermore, the computational performance of the frequency analysis, extracting hundreds of sinusoids from Kepler light curves in a few hours, demonstrates its value as a tool for a field like asteroseismology.

Multi-Spectral Food Classification and Caloric Estimation Using Predicted Images.

In nutrition science, methods that accomplish continuous recognition of ingested foods with minimal user intervention have great utility. Our recent study showed that using images taken at a variety of wavelengths, including ultraviolet (UV) and near-infrared (NIR) bands, improves the accuracy of food classification and caloric estimation. With this approach, however, analysis time increases as the number of wavelengths increases, and there are practical implementation issues associated with a large number of light sources. To alleviate these problems, we proposed a method that used only standard red-green-blue (RGB) images to achieve performance that approximates the use of multi-wavelength images. This method used RGB images to predict the images at each wavelength (including UV and NIR bands), instead of using the images actually acquired with a camera. Deep neural networks (DNN) were used to predict the images at each wavelength from the RGB images. To validate the effectiveness of the proposed method, feasibility tests were carried out on 101 foods. The experimental results showed maximum recognition rates of 99.45 and 98.24% using the actual and predicted images, respectively. Those rates were significantly higher than using only the RGB images, which returned a recognition rate of only 86.3%. For caloric estimation, the minimum values for mean absolute percentage error (MAPE) were 11.67 and 12.13 when using the actual and predicted images, respectively. These results confirmed that the use of RGB images alone achieves performance that is similar to multi-wavelength imaging techniques.

Fire Fighting Robot with Human Detection and Audio Recognition

This paper presents innovative approaches to firefighting robots, addressing critical challenges in fire safety. CeaseFire, equipped with water and CO2 sprays, ensures real-time operations prioritizing human safety, offering user-friendly handling and extended control range. The DaNI(sbRIO)-powered robot excels in rapid fire detection and extinguishing, providing a reliable solution for hazardous situations. In parallel, the YOLOv4-tiny implemented on a limited hardware robot enhances accuracy in victim rescue during firefighting missions. The Arduino UNO-based mini fire-fighting robot autonomously detects and extinguishes fires, minimizing risks to firefighters and incorporating user intervention through an Android app. Additionally, a modular robot employs computer vision, temperature, and UV-TRON sensors for enhanced fire detection, pathfinding, and extinguishing, promising a significant advancement in autonomous firefighting technology. AFDER, an Autonomous Fire Detecting and Extinguishing Robot, integrates robotics, computer vision, and firefighting tech, autonomously navigating with lidar-based mapping and obstacle avoidance. With precise fire detection and targeted extinguishing capabilities, AFDER reduces human intervention, proving adaptable to diverse environments. Collectively, these advancements showcase the potential to revolutionize firefighting, safeguard lives, and address the complex challenges posed by fire emergencies.

Energy-Efficient Implementation of the Lattice Boltzmann Method

Energy costs are now one of the leading criteria when procuring new computing hardware. Until recently, developers and users focused only on pure performance in terms of time-to-solution. Recent advances in energy-aware runtime systems render the optimization of both runtime and energy-to-solution possible by including hardware tuning depending on the application’s workload. This work presents the impact that energy-sensitive tuning strategies have on a state-of-the-art high-performance computing code based on the lattice Boltzmann approach called waLBerla. We evaluate both CPU-only and GPU-accelerated supercomputers. This paper demonstrates that, with little user intervention, when using the energy-efficient runtime system called MERIC, it is possible to save a significant amount of energy while maintaining performance.

Enabling Bitwise Reproducibility for the Unstructured Computational Motif

In this paper we identify the causes of numerical non-reproducibility in the unstructured mesh computational motif, a class of algorithms commonly used for the solution of PDEs. We introduce a number of parallel and distributed algorithms to address nondeterminism in the order of floating-point computations, in particular, a new graph coloring scheme that produces identical coloring results regardless of how many parts the graph is partitioned to. We implement these in the OP2 domain specific language (DSL) and show how it can be automatically deployed to any application that uses OP2 without user intervention. We contrast differences in results without reproducibility and then demonstrate how bitwise reproducibility can be gained using our methods on a variety of applications including a production CFD application used at Rolls-Royce. We evaluate the performance and overheads of enforcing bitwise reproducibility on a cluster of CPUs and GPUs.

Spectral Shifted Stabilized Continuation for Indirect Optimal Control

Continuation methods provide a solution to the initial guess problem that numerical root solvers and shooting methods often face by iteratively transforming an auxiliary solution into a desired solution through differential corrections. While seemingly straightforward, designing an automated continuation strategy that starts at the auxiliary problem and successfully terminates at the desired problem with minimal user intervention is often challenging especially for optimal trajectory-planning problems with nonlinear motion models subject to state/path and control constraints. One scenario where the auxiliary problem does not connect to the desired problem occurs when the intermediate problems are ill posed (i.e., infeasible). This paper presents a two-layered Jacobian conditioned stabilized continuation algorithm that circumvents these infeasible zones along a user-defined continuation path with little designer intervention. The efficacy of this approach is evaluated in the context of several prototype problems including quadratic root-solving, optimal path planning with a Dubins model, an optimal orbit transfer problem, and an optimal high-speed vehicle trajectory generation problem.

Hardware implementation of WPT Charging System for EV Vehicles

Transportation in urban areas is being transformed by various vehicles, with e-scooters being among the fastest growing. Despite their popularity, e-scooters face issues like incompatible chargers, especially problematic for rental services. Wireless charging emerges as a solution by enabling battery charging without user intervention. This paper focuses on the design and development of a magnetic-resonance charger for e-scooters. This research has detailed the coil topology, gap definition, and optimized control for a constant current-constant voltage (CC-CV) charge. This present key contribution is the integrated consideration of these factors, alongside the vehicle’s materials and structure, for precise design and implementation. The vehicle’s dimensions significantly constrain the coil design. Thus, in the past, a detailed analysis using Ansys Maxwell to determine the optimal locations for primary and secondary coils in an actual e-scooter was carried out. This analysis led to an optimal design for coil geometries, minimizing costs. The proposed system was validated with a real prototype, incorporating CC-CV control to ensure safe charging for various battery states, and is adaptable to a wide range of e-scooters, enhancing the usability of such chargers in public installations.

Estimation and Prediction of Glucose Appearance Rate for Use in a Fully Closed-Loop Dual-Hormone Intraperitoneal Artificial Pancreas.

A fully automated artificial pancreas requires a meal estimator and predictions of blood glucose levels (BGL) to handle disturbances during meal times, all without relying on manual meal announcements and user interventions. This study introduces a technique for estimating the glucose appearance rate (GAR) and predicting BGL in people with type 1 diabetes and insulin and glucagon administration. It is demonstrated for intraperitoneal insulin and glucagon delivery but may be adapted to other delivery sites. The estimator is designed based on the moving horizon estimation (MHE) approach, where the underlying cost function incorporates prior statistical information on the GAR in subjects over the course of a day. The proposed prediction scheme is developed to predict GAR using estimated states and an intestinal model, which is then used to predict BGL with the help of an animal glucose metabolic model. The intraperitoneal dual-hormone estimator was evaluated on three anesthetized animals, achieving a 21.8% mean absolute percentage error (MAPE) for GAR estimation and a 10.0% MAPE for BGL prediction when the future GAR is known. For a 120-minute prediction horizon, the proposed predictor achieved an 18.0% MAPE for GAR and a 28.4% MAPE for BGL. The findings demonstrate the effectiveness and reliability of the proposed estimator and its potential for use in a fully automated artificial pancreas and reducing user interventions. This study represents advancements toward the development of a fully automated artificial pancreas, ultimately enhancing the quality of life for people with type 1 diabetes.

DSIL4IMR: A dual step-based initial localization method for industrial mobile robot

This paper introduces a dual-step based initial localization method for industrial mobile robots (DSIL4IMR) that operates without user intervention at start-up. Utilizing a differential chassis equipped with a radar sensor, the method combines feature extraction and spatial information matching steps to locate the robot’s initial position. A unique metric, degree of stability, is developed to identify a feature map whether it is a dynamic or erroneous feature. DSIL4IMR does not require artificial landmarks and outperforms existing WiFi fingerprinting and map filter methods in both accuracy and efficiency. DSIL4IMR triples the localization efficiency compared to traditional approaches while achieving centimetre-level precision.

Fusing Utility and Interest: Recommendation of Healthcare Interventions for Online Users

Fusing Utility and Interest: Recommendation of Healthcare Interventions for Online Users