Archive ouverte HAL-LAAS
Not a member yet
12189 research outputs found
Sort by
Hybrid and networked dynamical systems: Modeling, Analysis and Control
No full textInternational audienceHybrid and Networked Dynamical Systems treats a class of systems that is ubiquitous in everyday life. From energy grids to fleets of robots or vehicles to social networks to biological networks, the same scenario arises: dynamical units interact locally through a connection graph to achieve a global task. The book shows how analysis and design tools can be adapted for control applications that combine the effects of network-induced interactions and hybrid dynamics with complex results.Following a scene-setting introduction, the remaining 12 chapters of the book are divided into three parts and provide a unique opportunity to describe the big picture that is the culmination of years of recent research activity. The contributing authors expand on their ideas at greater length than is possible in an archival research paper and use in-depth examples to illustrate their theoretical work.The widespread importance of hybrid and networked systems means that the book is of significant interest to academic researchers working in applied mathematics, control, and electrical, mechanical and chemical engineering and to their industrial counterparts
Activationless Electron Transfer of Redox-DNA in Electrochemical Nanogaps
No full textInternational audienceOur recent discovery of decreased reorganization energy in electrode-tethered redox-DNA systems prompts inquiries into the origin of this phenomenon and suggests its potential use to lower the activation energy of electrochemical reactions. Here, we show that the confinement of the DNA chain in a nanogap amplifies this effect to an extent to which it nearly abolishes the intrinsic activation energy of electron transfer. Employing electrochemical atomic force microscopy (AFM-SECM), we create sub-10 nm nanogaps between a planar electrode surface bearing end-anchored ferrocenylated DNA chains and an incoming microelectrode tip. The redox cycling of the DNA’s ferrocenyl (Fc) moiety between the surface and the tip generates a measurable current at the scale of ∼10 molecules. Our experimental findings are rigorously interpreted through theoretical modeling and original molecular dynamics simulations (Q-Biol code). Several intriguing findings emerge from our investigation: (i) The electron transport resulting from DNA dynamics is many times faster than predicted by simple diffusion considerations. (ii) The current in the nanogap is solely governed by the electron transfer rate at the electrodes. (iii) This rate rapidly saturates as overpotentials applied to the nanogap electrodes increase, implying near-complete suppression of the reorganization energy for the oxidation/reduction of the Fc heads within confined DNA. Furthermore, evidence is presented that this may constitute a general, previously unforeseen, behavior of redox polymer chains in electrochemical nanogaps
An experimental approach for evaluating cache allocation policies in multicore for real-time mixed-criticality systems
No full textInternational audienceEmbedded systems are experiencing an increasingdemand for computational power. A commonly adopted solutionto meet this demand involves deploying both critical and non-critical tasks on a single multicore processor. Nevertheless, theintricacy of such processors induces nondeterminism, posingpotential risks to the dependability of the system. This becomesparticularly pertinent in safety-critical real-time applicationswhere temporal faults could lead to missed deadlines for high-criticality tasks. Emerging technologies like Intel’s Cache Al-location Technology (CAT) are designed to diminish the non-determinism instigated by shared cache memory in multicoresystems, by enabling dynamic cache memory allocation. In thispaper, we introduce an experimental methodology to gauge theefficacy of such technology in a real-time setting. We investigatethe possibility of leveraging dynamic cache memory allocation toensure high-criticality tasks meet their deadlines while optimizingthe performance of non-critical tasks. Our proposed methodologyinvolves an exhaustive analysis of the trade-offs between variousparameters in a mixed-criticality application. The effectivenessof this approach is substantiated through a sensitivity analysison a practical use case
Parallel on-chip micropipettes enabling quantitative multiplexed characterization of vesicle mechanics and cell aggregates rheology
No full textInternational audienceMicropipette aspiration (MPA) is one of the gold standards for quantifying biological samples' mechanical properties, which are crucial from the cell membrane scale to the multicellular tissue. However, relying on the manipulation of individual home-made glass pipettes, MPA suffers from low throughput and no automation. Here, we introduce the sliding insert micropipette aspiration method, which permits parallelization and automation, thanks to the insertion of tubular pipettes, obtained by photolithography, within microfluidic channels. We show its application both at the lipid bilayer level, by probing vesicles to measure membrane bending and stretching moduli, and at the tissue level by quantifying the viscoelasticity of 3D cell aggregates. This approach opens the way to high-throughput, quantitative mechanical testing of many types of biological samples, from vesicles and individual cells to cell aggregates and explants, under dynamic physico-chemical stimuli
Design and fabrication of a fast-response and low-energy input micro igniter
No full textInternational audienceIn this study, we innovatively combine direct ink writing (DIW) and physical vapor deposition (PVD) to fabricate a novel micro igniter with fast-response and low-energy input characteristics. The igniter comprises a 5×5 mm² metal film bridge with gold contact pads. The metallic resistance is fabricated using a series of photolithography, metal deposition, and lift-off processes. After evaluating titanium and chromium as the resistive filament, titanium was selected due to its superior reactivity when in contact with CuO leading to reduced ignition energy. Onto the resistive titanium layer, either thermite multilayer films and/or energetic inks are deposited to achieve a head-to-head comparison of ignition characteristics. We show that igniters powered by (CuO/Ti)5 bilayers and Al/CuO/PVDF as energetic ink demonstrate the equivalent ignition performance as reactive (CuO/Ti)5 multilayered igniter. The ignition delay and ignition energy are below 1 ms and 2 mJ, respectively, while the flash intensity is approximately one decade higher than what is typically achieved with conventional micro-igniters. Furthermore, its ignition behaviour can be readily adjusted by altering the constituents of the fuel/oxidizer or adjusting the ratio of multilayer/inks, thereby allowing for easy adaptation of the micro-igniter to meet the diverse requirements of various applications
Rétrofit ALCATEL AMS 4200
No full textNational audienceThis document is a presentation of the retrofit carried out on the LAAS CNRS etching machine, an Alcatel AMS 4200.Ce document est une présentation du rétrofit qui a été réalisé sur la machine de gravure profonde du LAAS CNRS, une Alcatel AMS 4200
Plan Generation for Multi-Robot Missions Requiring Active Operator Involvement
No full textPapier présenté au workshop PlanROB d'ICAPS 2024, il présente une extension d'une approche de planification pour les missions multi-robots, en tenant compte de la charge cognitive des opérateurs. Des métriques de charge mentale sont intégrées pour optimiser la qualité des plans, avec une évaluation sur des scénarios réalistes.This paper extends a constraint-based planning approach to deal with mixed-initiative for complex multi-robot missions. Operators in the loop with multi-robot systems may have to interact intensively: explicitly considering their cognitive load while planning the missions is a critical problem to address. The purpose of this work is to take into account at mission planning time the operators capacity to supervise the mission execution, to ensure efficient and safe operations. We introduce new mental load related metrics in an automatic constraint-based planner. The optimization of these metrics yields better quality plans for the operators to supervise and interact during execution. The planning feasibility and performances are evaluated on realistic scenarios
Model predictive control under hard collision avoidance constraints for a robotic arm
No full textInternational audienceWe design a method to control the motion of a manipulator robot while strictly enforcing collision avoidance in a dynamic obstacle field. We rely on model predictive control while formulating collision avoidance as a hard constraint. We express the constraint as the requirement for a signed distance function to be positive between pairs of strictly convex objects. Among various formulations, we provide a suitable definition for this signed distance and for the analytical derivatives needed by the numerical solver to enforce the constraint. The method is completely implemented on a manipulator "Panda" robot, and the efficient open-source implementation is provided along with the paper. We experimentally demonstrate the efficiency of our approach by performing dynamic tasks in an obstacle field while reacting to non-modeled perturbations
Explaining the Space of SSP Policies via Policy-Property Dependencies: Complexity, Algorithms, and Relation to Multi-Objective Planning
No full textInternational audienceStochastic shortest path (SSP) problems are a common framework for planning under uncertainty. However, the reactive structure of their solution policies is typically not easily comprehensible by an end-user, nor do planners justify the reasons behind their choice of a particular policy over others. To strengthen confidence in the planner’s decision-making, recent work in classical planning has introduced a framework for explaining to the user the possible solution space in terms of necessary trade-offs between user-provided plan properties. Here, we extend this framework to SSPs. We introduce a notion of policy properties taking into account action-outcome uncertainty. We analyze formally the computational problem of identifying the exclusion relationships between policy properties, showing that this problem is in fact harder than SSP planning in a complexity theoretical sense. We show that all the relationships can be identified through a series of heuristic searches, which, if ordered in a clever way, yields an anytime algorithm. Further, we introduce an alternative method, which leverages a connection to multi-objective probabilistic planning to move all the computational burden to a preprocessing step. Finally, we explore empirically the feasibility of the proposed explanation methodology on a range of adapted IPPC benchmarks
Neuro-Symbolic Learning of Lifted Action Models from Visual Traces
No full textInternational audienceModel-based planners rely on action models to describe available actions in terms of their preconditions and effects. Yet, manually encoding such models is challenging, especially in complex domains. Numerous methods have been proposed to learn action models from examples of plan execution traces. However, high-level information, such as state labels within traces, is often unavailable and needs to be inferred indirectly from raw observations. In this paper, we aim to learn lifted action models from visual traces — sequences of image-action pairs depicting discrete successive trace steps. We present ROSAME, a differentiable neuRO-Symbolic Action Model lEarner that infers action models from traces consisting of probabilistic state predictions and actions. By combining ROSAME with a deep learning computer vision model, we create an end-to-end framework that jointly learns state predictions from images and infers symbolic action models. Experimental results demonstrate that our method succeeds in both tasks, using different visual state representations, with the learned action models often matching or even surpassing those created by humans