HAL-CentraleSupelec
Not a member yet
77624 research outputs found
Sort by
Sex Education Programs’ Content Analysis and Perception Among Spinal Cord Injury Individuals: A Scoping Review
No full textInternational audiencePeople living with Spinal Cord Injury (pwSCI) are likely to experience sexuality issues. While sex education is an option to help manage these challenges, this scoping review aimed to identify the content of sex education interventions provided to pwSCI and to analyze participants' perceptions of these interventions. A scoping review was performed in PubMed, Cochrane library, Science Direct, PEDro and Web of Science including SciELO until January 30, 2025. Studies were included if patients received a sex education intervention and if a content analysis or patient opinion was available. Two reviewers conducted the study selection, and a third reviewer resolved conflicts. Results: The search yielded 3449 records, from which 18 studies were included. 14 studies reported formal education programs, while 4 studies focused on the variety of sex education experiences pwSCI received as part of rehabilitation. Program modalities varied, but shared similarities: mainly face-to-face delivery, post-rehabilitation care, external resources provision, delivered either in a one-time intensive session or over several weeks, mainly by physicians, nurses and psychologists. Topics were both physical and psychological, covering physiology as well as relationships and self-esteem, and suggesting practical strategies. Intervention perception findings suggest that formal sex education programs meet the needs of pwSCI with overall majority being satisfied and recommending these programs to other pwSCI. Information provided during routine care appears insufficient and divergent from pwSCI expectations. This study highlights the importance of designing sex education programs tailored to the specific needs of pwSCI targeting primary, secondary and tertiary sexual dysfunctions
CODE beyond FAIR: a roadmap for reusable research software
No full textInternational audienceFAIR principles are a set of guidelines aiming at simplifying the distribution of scientific data to enhance reuse and reproducibility. This article focuses on research software, which significantly differs from data in its living nature, and its relationship with free and open-source software. We provide a tiered roadmap to improve the state of research software, which takes into account the full range of stakeholders in the research software ecosystem: all scientific staff – regardless of prior software engineering training – but also institutions, funders, libraries and publishers
A hybrid urban delivery system with robots
No full textInternational audienceCities are now restricting access for conventional delivery technologies in some areas, requiring businesses to adopt more flexible distribution systems to complete their deliveries. We present a two-echelon hybrid truck-based robot delivery system for last-mile logistics operations. Robots can navigate through truck no-go areas such as pedestrian zones and college campuses, while trucks can travel through less restricted areas. The hybrid delivery model allows the distribution system to automatically select the better distribution strategy, thereby improving distribution efficiency.We present a mixed-integer linear program to model the proposed system. We also offer valid inequalities to strengthen the formulation and a large neighborhood search-based algorithm with innovative adaptive methods and multiple operators to solve medium and large-scale instances efficiently. Computational experiments are conducted to evaluate how our proposed model performs. Sensitivity analysis experiments considering truck no-go areas of different sizes and area access time windows are performed and reveal managerial insights. We suggest setting up appropriate time windows for some truck no-go areas to reduce the burden of logistics companies in increasing distribution routes due to access restrictions
Parameter influence analysis in a 3D TBM model via sensitivity analysis and GP metamodels
No full textInternational audienceUrban tunnel excavation with tunnel boring machines induces ground movements that can affect nearby structures. Three-dimensional finite element models (FEM) are widely used to predict these settlements, but their high computational cost limits direct exploration of parameter influence. This work presents a 3D FEM simulator of mechanized tunneling and a methodology to quantify the impact of both numerical and physical inputs on settlement predictions. First, an accuracy-cost model reduction study evaluates the effect of domain dimensions and mesh densities on a small number of scalar quantities of interest extracted from simulated settlement fields. Empirical error models are fitted and used to select a reduced configuration that balances accuracy and runtime. Second, Gaussian process models are trained on simulation data from the reduced configuration and validated using exact leave-one-out cross-validation. These metamodels enable the computation of Sobol’ sensitivity indices with quantified uncertainty, identifying the most influential geological, operational, and loading parameters. The proposed framework reduces the cost of sensitivity analysis for computationally intensive 3D tunneling simulations, supporting input screening and dimensionality reduction for design and calibration
Cramér–Rao bound analysis of nested arrays under impulsive noise with coarrays and FLOSs
No full textInternational audienceNested arrays are extensively employed in array signal processing to augment the degrees of freedom and enhance estimation precision, and the Cramér–Rao Bound (CRB) for nested arrays in a Gaussian noise environment has been established. Nevertheless, in a practical wireless communication environment, noise usually exhibits an impulsive characteristic. The impulsive noise applied in a uniform linear array (ULA) has been extensively studied in the literature, but only closed-form expressions of CRB with Cauchy and Gaussian noise distributions are given. Although nested arrays have garnered significant attention recently, research on the CRB under impulsive noise conditions remains scarce. In this paper, we provide the CRB expression for direction of arrival (DOA) estimation with nested arrays in an impulsive noise environment, which indicates that the CRB is formulated in terms of the fractional low-order statistics (FLOSs) of received data. Moreover, we also calculate the CRB results for different FLOSs as well as various sparse arrays and validate that the derived CRB makes an important contribution to the performance analysis of sparse arrays in impulsive noise
Explicit Abstraction Barrier for Autoactive Verification
No full textInternational audienceComa is a verification language that allows the programmer to decide which part of a function implementation is visible to (and verified by) the caller, and which part is hidden from the caller and verified at the definition site.In this paper, we show through a series of examples how this functionality allows for extra flexibility, leading to more concise and natural specifications—if we write them at all
Robust Detection of Synthetic Tabular Data under Schema Variability
No full textInternational audienceThe rise of powerful generative models has sparked concerns over data authenticity. While detection methods have been extensively developed for images and text, the case of tabular data, despite its ubiquity, has been largely overlooked. Yet, detecting synthetic tabular data is especially challenging due to its heterogeneous structure and unseen formats at test time. We address the underexplored task of detecting synthetic tabular data "in the wild'', i.e. when the detector is deployed on tables with variable and previously unseen schemas. We introduce a novel datum-wise transformer architecture that significantly outperforms the only previously published baseline, improving both AUC and accuracy by 7 points. By incorporating a table-adaptation component, our model gains an additional 7 accuracy points, demonstrating enhanced robustness. This work provides the first strong evidence that detecting synthetic tabular data in real-world conditions is feasible, and demonstrates substantial improvements over previous approaches. Following acceptance of the paper, we are finalizing the administrative and licensing procedures necessary for releasing the source code. This extended version will be updated as soon as the release is complete
EMG-based Torque Prediction for Assistive Exoskeleton Control using Neural Networks with Bounded Generalization Error
No full textElectromyography (EMG) signals are widely used in assistive exoskeleton control for predicting human joint torque due to their ability to extract muscle activations before movement onset. The standard procedure for learning the EMG-to-torque model involves training the model on a training set of EMG-torque data, followed by validating the model on a separate test set. The comparison between models is generally undertaken on the test set. However, the analysis of model performance on the data outside the test set remains unaddressed. The lack of a guarantee for unseen data reduces the reliability of EMG-to-torque models in practical exoskeleton control. In this paper, we address this issue by proposing a bounded-generalization-error neural network (BGNN) for EMG-based torque prediction. Using gradient descent to train the network, we formulate at each training step a theoretical upper bound on the generalization error, reflecting the prediction error across the entire data distribution, including unseen data beyond the test set. The NN training is terminated when this upper bound reaches its minimum, thereby achieving the tightest guarantee on the generalization error. Experimental results on torque prediction demonstrated that, while ensuring such a bounded generalization error, our method still gave results comparable to those of classical models. The use of our BGNN in assistive exoskeleton control was also tested with 13 participants on a pick-and-place task with an upper limb exoskeleton. Experimental results on assistive control revealed that our method can reduce human physical fatigue without compromising movement speed or accuracy compared to natural human movement characteristics, particularly for generalization in novel tasks
Mitigating Systematic Errors in Roadside Camera-Based V2X Perception
No full textInternational audienceAs Cooperative Intelligent Transport Systems (C-ITS) continue to evolve, roadside infrastructure equipped with cameras, lidars, and radars to detect and report objects through Collective Perception Messages (CPMs) will play a crucial role in enhancing the situational awareness of connected users within the monitored environment. In the longer term, Connected and Autonomous Vehicles (CAVs) are expected to integrate such information with their on-board sensor data to support decisionmaking processes that may directly impact road users safety. To enable this, infrastructure-generated messages must provide highly accurate information on detected objects, particularly regarding their kinematic properties and associated uncertainties. However, camera sensors are sensitive to environmental factors such as object distance, position within the Field of View (FoV), weather, and lighting conditions. In this work, we propose a method to dynamically estimate and mitigate systematic errors in camera-based kinematic state estimation by an infrastructure. The correction mechanism relies on an initial offline deviation analysis performed using a probe vehicle equipped with highprecision instruments that moves through the camera's FoV. We evaluate the method under diverse weather and lighting conditions and demonstrate a significant reduction in kinematic estimation deviations after correction
DOA Estimation by Reduced-Dimension MUSIC with Uniform Circular Array
No full textInternational audienceUniform circular array (UCA) is widely employed in the field of direction-of-arrival (DOA) estimation, due to its uniform coverage in all directions and its ability to estimate both azimuth and elevation angles. Conventional methods, such as two-dimensional MUSIC (2D-MUSIC) or beam-space transform based methods, either require computationally intensive multi-dimensional searches, or suffer from residual error and bias in DOA estimation, which limit their practicality in realtime applications. To address these challenges, we devise a reduced-dimension MUSIC (RD-MUSIC) for DOA estimation by investigating both odd and even numbers of UCA sensors. The proposed method rearranges the sensor positions, and then splits the steering vector into two components, allowing for separate 1D searches for the elevation and azimuth angles. Our solution can achieve high estimation accuracy in a computationally attractive manner. The effectiveness of the RD-MUSIC is evaluated through numerical simulations as well as laboratory and field experiments, demonstrating its robustness at different signal-to-noise ratio levels and number of snapshots. The results highlight the potential of RD-MUSIC for real-time applications where computational resources are limited and high accuracy is required