HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Preventive Time Slot Allocation MAC Protocol for Vehicular Networks
No full textInternational audienc
Cooperative Architecture Using Air and Ground Vehicles for the Search and Recognition of Targets
No full textInternational audienceA cooperative navigation architecture for the search and recognition of targets using aerial and ground vehicles is proposed in this paper. The architecture allows to manage aerial and ground vehicles to autonomously perform different tasks in an independent or cooperative way. For our application, two main tasks are conceived; aerial monitoring of a surface to search for targets, and target ground recognition. In the target aerial detection task, the aerial drone tracks autonomously a trajectory, computed to cover all the surface to monitoring, to search for targets using vision algorithms. Once one of them is detected its relative position is sent to the cooperative architecture. After the aerial drone has covered the entire area, the architecture computes and assigns to each ground vehicle the closest target found. Then, each ground vehicle navigates autonomously avoiding obstacles (if presents) to its assigned target. For verifying the success of the mission, the aerial vehicle flies following the dynamic center of mass of the ground vehicles. Real-time experiments are carried out to validate the proposed architecture. Main results, depicted in some graphs, corroborate the good performance in closed loop
More powerful energetic reasoning for the Cumulative Scheduling Problem
No full textInternational audienc
Sequential vs. integrated model of process planning, layout and scheduling optimization for RMS
No full textInternational audienceCompanies are currently facing the challenge of adapting to market changes, leading to the emergence of new production systems, including reconfigurable manufacturing systems (RMS). Another question to the classical resource allocation and scheduling decisions is when and how to reconfigure. This work proposes to answer all these questions by presenting two models to solve process planning, layout, and scheduling problems. Some examples were presented and solved by constraint programming to analyze the difference in results and execution times of the approaches
Adsorptive Performance of Walnut Shells Modified with Urea and Surfactant for Cationic Dye Removal
No full textInternational audienceAdsorption of cationic dye crystal violet (CV) onto a modified walnut shell (WS) adsorbent was investigated. Combined treatment of WS using urea and sodium dodecylsulfate (SDS) was carried out. Surface modification of adsorbents was confirmed by FTIR analysis, pHpzc measurements, and elemental and SEM-EDX analysis. In order to optimize the adsorption conditions, the effect of solution pH, adsorbent dose and CV concentration was studied by means of central composite face-centered design (CCFD). The highest correlation between experimental and model data was obtained for the pseudo-second-order (PSO) kinetic model, assuming an ion exchange mechanism of adsorption. A satisfactory fit of CV adsorption data was obtained from the Langmuir adsorption isotherm, supporting a single layer adsorption. According to obtained results, modified WS can be considered as a low-cost, efficient and environmentally compatible biosorbent for the removal of cationic pollutants from aqueous solutions
Méthodes de segmentation des voies biliaires sous 3D Slicer appliquées à la CPRE : avantages et inconvénients
No full textInternational audienceThis article presents an evaluation of biliary tract segmentation methods used for 3D reconstruction, which may be very usefull in various critical interventions, such as endoscopic retrograde cholangiopancreatography (ERCP), using the 3D Slicer software. This article provides an assessment of biliary tract segmentation techniques employed for 3D reconstruction, which can prove highly valuable in diverse critical procedures like endoscopic retrograde cholangiopancreatography (ERCP) through the utilization of 3D Slicer software. Three different methods, namely thresholding, flood filling, and region growing, were assessed in terms of their advantages and disadvantages. The study involved 10 patient cases and employed quantitative indices and qualitative evaluation to assess the segmentations obtained by the different segmentation methods against ground truth. The results indicate that the thresholding method is almost manual and time-consuming, while the flood filling method is semi-automatic and also time-consuming. Although both methods improve segmentation quality, they are not reproducible. Therefore, an automatic method based on region growing was developed to reduce segmentation time, albeit at the expense of quality. These findings highlight the pros and cons of different conventional segmentation methods and underscore the need to explore alternative approaches, such as deep learning, to optimize biliary tract segmentation in the context of ERCP
A Deep Reinforcement Learning Decision-Making Approach for Adaptive Cruise Control in Autonomous Vehicles
No full textInternational audienceIn the evolving automobile industry, Adaptive Cruise Control (ACC) is key for aiding autonomous traffic navigation. Ideal ACC systems can decelerate to low speeds in stop-and-go traffic, maintain a safe following distance, minimize rear-end collision risks, and lessen the driver's need to continually adjust vehicle's speed to match traffic flow. In this paper, we offer a Deep Reinforcement Learning-based adaptive cruise control (DRL-ACC) system that creates safe, flexible, and responsive car-following policies agents. Instead of using discrete incremental and decremental values or a continuous action space, we suggest constructing a discrete high-level action space to accelerate, decelerate, and hold the current speed. We also provide a comprehensive, easyto-interpret multi-objective reward function that reflects safe, responsive, and rational traffic behavior. This strategy, trained on a single steady-state flow car-following scenario, promotes steadiness, responsiveness, and shows better generalization to diverse car-following scenarios. Results are also compared to the conventional Intelligent Driver Model (IDM). We further explore the model's potential to avoid rear-end collisions and facilitate future integration of lane-change maneuvers, which will increase its effectiveness in emergency situations
On-ramp Merging on Highway for Cooperative Automated Vehicles based on an Online Reconfigurable Formation Control Approach
No full textInternational audienceOn-ramp merging scenarios remain among the most complex challenges for Autonomous Vehicle (AV) technology despite the significant advancements. In this paper, instead of considering individually each AV during the merging, it is proposed to take advantage from the Cooperative Automated Vehicles (CAVs) to tackle the on-ramp merging on highway. The main contribution of this paper is an overall cooperation strategy and formation control approach based on the online cooperative formation reconfiguration strategy, called Formation Reconfiguration Approach based on an Online Control Strategy (FRA-OCS). The proposed strategy operates under the cooperative mode part of the Altruistic Formation Reconfiguration Strategy (AFRS) [2]. To overcome the limitations of both the Constrained Optimal Reconfiguration Matrix (CORM) [1], and the Extended Constrained Optimal Reconfiguration Matrix (E-CORM) [2], the proposed FRA-OCS extends its functionality to ensure both formation safety criteria and efficient formation reconfiguration during the merging maneuvers, allowing thus even more reliable and flexible CAVs coordination. Several simulations are performed to evaluate the safety and reliability of the proposed approach
3D reconstruction of a horse swimming
No full text3D reconstruction of a horse swimming and visualisation of the joint angles during two cycles of swimming
Generation of β-like cell subtypes from differentiated human induced pluripotent stem cells in 3D spheroids
No full textSequencing data supporting the findings presented in this study were deposited at Zenodo (https://zenodo.org) with the following Digital Object Identifier: 10.5281/zenodo.7960673. All supplementary table data files and supplementary figures prepared for this study are available in the ESI supplementary files provided to the journal.International audienceSince the identification of four different pancreatic β-cell subtypes and bi-hormomal cells playing a role in the diabetes pathogenesis, the search for in vitro models that mimics such cells heterogeneity became a key priority in experimental and clinical diabetology. We investigated the potential of human induced pluripotent stem cells to lead to the development of the different β-cells subtypes in honeycomb microwell-based 3D spheroids. The glucose-stimulated insulin secretion confirmed the spheroids functionality. Then, we performed a single cell RNA sequencing of the spheroids. Using a knowledge-based analysis with a stringency on the pancreatic markers, we extracted the β-cells INS+/UCN3+ subtype (11%; β1-like cells), the INS+/ST8SIA1+/CD9− subtype (3%, β3-like cells) and INS+/CD9+/ST8SIA1-subtype (1%; β2-like cells) consistently with literature findings. We did not detect the INS+/ST8SIA1+/CD9+ cells (β4-like cells). Then, we also identified four bi-hormonal cells subpopulations including δ-like cells (INS+/SST+, 6%), γ-like cells (INS+/PPY+, 3%), α-like-cells (INS+/GCG+, 6%) and ε-like-cells (INS+/GHRL+, 2%). Using data-driven clustering, we extracted four progenitors’ subpopulations (with the lower level of INS gene) that included one population highly expressing inhibin genes (INHBA+/INHBB+), one population highly expressing KCNJ3+/TPH1+, one population expressing hepatocyte-like lineage markers (HNF1A+/AFP+), and one population expressing stem-like cell pancreatic progenitor markers (SOX2+/NEUROG3+). Furthermore, among the cycling population we found a large number of REST+ cells and CD9+ cells (CD9+/SPARC+/REST+). Our data confirm that our differentiation leads to large β-cell heterogeneity, which can be used for investigating β-cells plasticity under physiological and pathophysiological conditions