INRIA a CCSD electronic archive server
Not a member yet
122212 research outputs found
Sort by
Faciliter les collaborations humains-robots en téléopération haptique, des approches de contrôle en autonomie partagée
No full textShared control frameworks assist human operators by blending their commands with autonomous, goal-oriented trajectories. However, conventional blending techniques often fail to guarantee the feasibility of the resulting motion or the optimality of the combined decision. This thesis addresses two principal gaps in shared control: 1) the lack of a blending arbitrator that unifies predictive foresight with verifiable safety in a computationally tractable manner, and 2) the flawed assumption that the autonomous assistance is correct, which leads to performance degradation and user-robot conflict when the system’s world model is misaligned with reality. This work presents a control architecture that resolves both challenges. First, to address the arbitration gap, we formulate blending as a constrained optimal control problem. A Model Predictive Control for Blending (MPC-B) framework is proposed to compute a feasible blended trajectory via receding-horizon optimization, ensuring proactive compliance with all system and task constraints. Second, to address the assistance gap, we introduce a Dual-Component Adaptive Assistance Framework that corrects for model inaccuracies by treating the operator’s input as a corrective measurement. This framework integrates a real-time Adaptive Kalman Filter to compensate for local, transient errors and an online N-Point Procrustes Analysis module to learn and correct for global, systematic misalignments over time. The proposed architecture is evaluated in two human-in-the-loop teleoperation studies. The experimental results demonstrate the superiority of the proposed frameworks compared to conventional blending and unassisted teleoperation. The MPC-B controller significantly improved safety by eliminating kinematic constraint violations, while the adaptive assistance framework successfully overcame significant model errors to improve task efficiency beyond unassisted capabilities. Taken together, the results validate the integrated architecture as a solution for safer and more effective human-robot collaboration, yielding quantifiable improvements in task performance, interaction quality, and operator workload.Les cadres de contrôle partagé assistent les opérateurs humains en fusionnant leurs commandes avec des trajectoires autonomes orientées vers un objectif. Cependant, les techniques de fusion conventionnelles ne garantissent souvent ni la faisabilité du mouvement résultant, ni l’optimalité de la décision combinée. Cette thèse aborde deux lacunes principales du contrôle partagé : 1) l’absence d’un arbitre de fusion qui unifie la prévoyance prédictive avec une sécurité vérifiable d’une manière calculatoirement tractable, et 2) l’hypothèse erronée que l’assistance autonome est correcte, ce qui entraîne une dégradation des performances et un conflit entre l’utilisateur et le robot lorsque le modèle du monde du système est désaligné avec la réalité. Ce travail présente une architecture de contrôle holistique qui résout ces deux défis. Premièrement, pour combler la lacune de l’arbitrage, nous formulons la fusion comme un problème de contrôle optimal sous contraintes. Un cadre de Commande Prédictive pour la Fusion (MPC-B) est proposé pour calculer une trajectoire fusionnée réalisable via une optimisation à horizon fuyant, assurant une conformité proactive avec toutes les contraintes du système et de la tâche. Deuxièmement, pour combler la lacune de l’assistance, nous introduisons un Cadre d’Assistance Adaptative à Double Composante qui corrige les inexactitudes du modèle en traitant l’entrée de l’opérateur comme une mesure corrective. Ce cadre intègre un Filtre de Kalman Adaptatif en temps réel pour compenser les erreurs locales et transitoires et un module d’Analyse de Procrustes à N-points en ligne pour apprendre et corriger les désalignements globaux et systématiques au fil du temps. L’architecture proposée a été évaluée dans deux études complètes de téléopération avec des humains dans la boucle. Les résultats expérimentaux démontrent la supériorité des cadres proposés par rapport à la fusion conventionnelle et à la téléopération non assistée. Le contrôleur MPC-B a considérablement amélioré la sécurité en éliminant les violations de contraintes cinématiques, tandis que le cadre d’assistance adaptative a surmonté avec succès d’importantes erreurs de modèle pour améliorer l’efficacité de la tâche au-delà des capacités non assistées. Ensemble, les résultats valident l’architecture intégrée comme une solution robuste pour une collaboration homme-robot plus sûre et plus efficace, produisant des améliorations quantifiables en matière de performance de la tâche, de qualité de l’interaction et de charge de travail de l’opérateur
Which theories have a measurement problem?
No full textInternational audienceAbstract It is shown that any theory that has certain properties has a measurement problem, in the sense that it makes predictions that are incompatible with measurement outcomes being absolute (that is, unique and non-relational). These properties are Bell Nonlocality, Information Preservation, and Local Dynamics. The result is extended by deriving Local Dynamics from No Superluminal Influences, Separable Dynamics, and Consistent Embeddings. These results are achieved using a framework of \textit{perspectival theories} that generalizes the notion of a Heisenberg cut beyond quantum theory. As well as explaining why the existing Wigner's-friend-inspired no-go theorems hold for quantum theory, the results also shed light on whether a future theory of physics might overcome the measurement problem. In particular, they suggest the possibility of a theory in which absoluteness is maintained, but without rejecting relativity theory (as in Bohm theory) or embracing objective collapses (as in GRW theory)
Assessing the impact of biomass retention in membrane-assisted microalgae-bacteria sewage treatment
No full textInternational audienceMicroalgae/bacteria consortia (MBC) are considered a promising platform for wastewater treatment. Biomass retention based on membrane filtration represents an effective alternative to enhance organic matter and nutrient removal. This study investigated the effects of uncoupling solid from hydraulic retention time (SRT and HRT, respectively) when treating synthetic municipal wastewater. For that purpose, 50 L high-rate algal pond was coupled with a membrane filtration step and operated at different conditions. The performance of the system was evaluated in terms of its load and concomitant removal of organic matter and nutrients. Organic matter removal exceeded 90 %. Nitrogen removal efficiencies were in the range 46-68 %, with the highest nitrogen removal rate exceeding 40 g N m⁻³ d⁻¹ at HRT of 1 day. Phosphorus removal efficiencies varied between 39 % and 70 %. High loads resulting from low HRTs caused low oxygen levels in the reactor, allowing favourable conditions for the occurrence of denitrification. Experimental data supported the possibility of having in the same treatment unit different nitrogen removal mechanisms: assimilation, nitrification, and denitrification.</div
Constrained Average-Reward Intermittently Observable MDPs
No full textInternational audienceIn Markov Decision Processes (MDPs) with intermittent state information, decision-making becomes challenging due to periods of missing observations. Linear programming (LP) methods can play a crucial role in solving MDPs, in particular, with constraints. However, the resultant belief MDPs lead to infinite dimensional LPs, even when the original MDP is with finite state and action spaces. The verification of strong duality becomes non-trivial. This paper investigates the conditions for no duality gap in average-reward finite Markov decision process with intermittent state observations. We first establish that in such MDPs, the belief MDP is unichain if the original Markov chain is recurrent. Furthermore, we establish strong duality of the problem, under the same assumption. Finally, we provide a wireless channel example, where the belief state depends on the last channel state received and the age of the channel state. Our numerical results indicate interesting properties of the solution
Cardiac-vagal rhythm echoes on the heartbeat’s mechanosensory imprint in the brain
No full textInternational audienceThe heart, a highly reactive and innervated organ, plays a crucial role in brain-viscera communications. Recent research has highlighted the role of mechanosensation in the brain, where ion channels in neurons’ membranes respond to heartbeat-induced pressure changes, triggering specific neural responses. Cardiac mechano-electric coupling ensures cardiac output to match venous return through beat-by-beat feedback. However, the effect of ongoing cardiac rhythms on the brain-sensed strength of each heartbeat is not well understood. This is crucial for exploring brain-heart communication pathways and for understanding the mutual influence between brain and cardiac oscillations. This study explores how cardiac rhythms influence heartbeat strength (HBS) as detected by the brain in humans. As a proxy for brain-sensed HBS, we used ballistocardiographs, which capture HBS from the back of the head while participants are in horizontal position. By modeling HBS, we demonstrate that fast fluctuations in heart rate variability significantly influences the final HBS. This suggests a direct relationship between vagal tone and subsequent neural responses to heartbeats, highlighting the necessity of studying visceral oscillations in the context of mechanosensation and inter-organ communication research
A comparative analysis of metamodels for 0D cardiovascular models, and pipeline for sensitivity analysis, parameter estimation, and uncertainty quantification
No full textInternational audienceZero-dimensional (0D) cardiovascular models are reduced-order models aimed at studying the global dynamics of the whole circulation system or transport within it. They are employed to obtain estimates of important biomarkers for surgery planning and assessment applications (such as pressures, volumes, flow rates, and concentrations in the circulation) and can provide boundary conditions for high-fidelity three-dimensional models. Despite their low computational cost, tasks such as parameter estimation or uncertainty quantification require a large number of model evaluations, which is still a computationally expensive task. This motivates the building of metamodels in an offline stage, which can be evaluated significantly faster than 0D models. In this work, a pipeline going from 0D cardiovascular models to the building of metamodels and showcasing their use for tasks such as sensitivity analysis, parameter estimation, or uncertainty quantification is proposed. Three different strategies are assessed to build metamodels for 0D cardiovascular models, namely Neural Networks, Polynomial Chaos Expansion, and Gaussian Processes. The metamodels are assessed for three different 0D models. The first is a lumped model aimed at predicting the pressure in the portal vein after surgery. Due to the strong interaction between local liver hemodynamics and global circulation, the full circulation is modeled. The second one is simulating the whole-body circulation under the conditions of pulmonary arterial hypertension before and after shunt insertion. The final model is aimed at assessing the blood perfusion of an organ after a revascularization surgery. The transport of a contrast agent is modeled on top of a simplified 0D hemodynamics model. This model is chosen due to the different nature of the output which is a signal (concentration of the contrast agent over time), which requires a different treatment from the metamodeling point of view. The metamodels are trained and tested on synthetic data generated from the 0D models. It was found that neural networks offer the most convenient way of building metamodels in terms of the quality of the results, computational time, and practical ease of performing parameter estimation, sensitivity analysis, or uncertainty quantification tasks. Finally, we demonstrate a full pipeline of sensitivity analysis, inverse problem and (patient-specific) UQ, with a neural network as emulator
The Ready-Mixed Concrete Dispatch Problem with Uncertain Delivery Times
No full textInternational audienc
An AutoML Framework using AutoGluonTS for Forecasting Seasonal Extreme Temperatures (Accepted in IJCNN'2025, Wokshop: AI for a Cooler Planet: Tackling Environmental Challenges with Neural Networks): Published in the Workshop "AI for a Cooler Planet: Tackling Environmental Challenges with Neural Networks", part of the International Joint Conference on Neural Networks (IJCNN) 2025
No full textInternational audienceIn recent years, great progress has been made in the field of forecasting meteorological variables. Recently, deep learning architectures have made a major breakthrough in forecasting the daily average temperature over a ten-day horizon. However, advances in forecasting events related to the maximum temperature over short horizons remain a challenge for the community. A problem that is even more complex consists in making predictions of the maximum daily temperatures in the short, medium, and long term. In this work, we focus on forecasting events related to the maximum daily temperature over mediumterm periods (90 days). Therefore, instead of addressing the problem from a meteorological point of view, this article tackles it from a climatological point of view. Due to the complexity of this problem, a common approach is to frame the study as a temporal classification problem with the classes: maximum temperature above normal, normal or below normal. From a practical point of view, we created a large historical dataset (from 1981 to 2018) collecting information from weather stations located in South America. In addition, we also integrated exogenous information from the Pacific, Atlantic, and Indian Ocean basins. We applied the AutoGluonTS platform to solve the above-mentioned problem. This AutoML tool shows competitive forecasting performance with respect to large operational platforms dedicated to tackling this climatological problem; but with a "relatively" low computational cost in terms of time and resources.</div