A5003467974- Robot Manipulation and Learning
- Robotic Locomotion and Control
- Prosthetics and Rehabilitation Robotics
- Robotic Path Planning Algorithms
- Robotic Mechanisms and Dynamics
- Ergonomics and Musculoskeletal Disorders
- Muscle activation and electromyography studies
- Control and Dynamics of Mobile Robots
- Teleoperation and Haptic Systems
- Musculoskeletal pain and rehabilitation
- Motor Control and Adaptation
- Reinforcement Learning in Robotics
- Ergonomics and Human Factors
- Human-Automation Interaction and Safety
- Modular Robots and Swarm Intelligence
- Soft Robotics and Applications
- Real-time simulation and control systems
- Advanced Control Systems Optimization
- Dynamics and Control of Mechanical Systems
- Manufacturing Process and Optimization
- Social Robot Interaction and HRI
- Muscle Physiology and Disorders
- Neurogenetic and Muscular Disorders Research
- Advanced Vision and Imaging
- Stroke Rehabilitation and Recovery
Université de Bordeaux
2019-2024
Institut national de recherche en informatique et en automatique
2020-2024
Institut Systèmes Intelligents et de Robotique
2012-2023
Centre de Recherche Inria Bordeaux - Sud-Ouest
2019-2023
Institut Polytechnique de Bordeaux
2019-2022
Centre National de la Recherche Scientifique
2012-2021
Laboratoire de l'Intégration du Matériau au Système
2021
Sorbonne Université
2011-2020
Tactus Verslavingszorg
2020
Université Paris Cité
2014-2017
We present a novel approach to deal with transitions while performing sequence of dynamic tasks humanoid robot. The simultaneous achievement several cannot be ensured, so we use strategy based on weights represent their relative importance. robot interacts changing environment, and the input torques are different depending whether performs in constrained state (e.g. contact) or not. develop solution smooth variations transitional which avoids sharp torque evolutions. In order validate this...
The number of tools for dynamics simulation has grown substantially in the last few years. Humanoid robots, particular, make extensive use such a variety applications, from simulating contacts to planning complex motions. It is necessary humanoid robotics community have systematic evaluation assist choosing which available best their research. This paper surveys state art and reports on analysis an online survey about research community. major requirements researchers are better physics...
This paper addresses the problem of active object learning by a humanoid child-like robot, using developmental approach. We propose cognitive architecture where visual representation objects is built incrementally through exploration. present design guidelines architecture, its main functionalities, and we outline process robot showing how it learns to recognize in human-robot interaction scenario inspired social parenting. The actively explores manipulation, driven combination guidance...
The number of tools for dynamics simulation has grown in the last years. It is necessary robotics community to have elements ponder which available best their research. As a complement an objective and quantitative comparison, difficult obtain since not all are open-source, element evaluation user feedback. With this goal mind, we created online survey about use dynamical robotics. This paper reports analysis participants' answers descriptive information fiche most relevant tools. We believe...
SUMMARY The work presented in this paper aims at providing a unified modelling framework for the reactive control of wheeled mobile manipulators (WMM). Where most literature often provides models, sometimes simplified, given type WMM, an extensive description obtaining explicit kinematic and dynamic models those systems is given. This particularly well suited approaches, which, case manipulation missions, are necessary to handle complexity tasks be fulfilled, aspect extended workspace...
Balance strategies range from continuous postural adjustments to discrete changes in contacts: their simultaneous execution is required maintain stability while considering the engaged walking activity. In order compute optimal time, duration and position of footsteps along with center mass trajectory a humanoid, novel mixed-integer model system presented. The introduction this predictive control problem brings definition Mixed-Integer Quadratic Program, subject linear constraints....
Predicting the post-impact velocity for torque-controlled flexible-joint robots enhances impact-aware control schemes which exploit intentional collisions achieving dynamic robotic manipulation and locomotion. Compared to a previous approach based on fully rigid-robot assumption, this paper shows how an improvement in prediction can be obtained by taking into account joints' motor inertias, transmission ratios, low-level torque gains, as well impact surface friction. The letter also proposes...
The success of robots in real-world environments is largely dependent on their ability to interact with both humans and said environment. FP7 EU project CoDyCo focused the latter these two challenges by exploiting rigid compliant contacts dynamics robot control problem. Regarding former, properly manage interaction side, an estimation human behaviors intentions necessary. In this letter, we present building blocks such a human-in-the-loop controller, validate them simulation iCub humanoid...
Whole-Body Control methods offer the potential to execute several tasks on highly redundant robots, such as humanoids. Unfortunately, task combinations often result in incompatibilities which generate undesirable behaviors. Prioritization techniques can prevent from perturbing one another but detriment of lower precedence tasks. For many tasks, static prioritization is not necessary or even appropriate because be achieved variable ways, reaching. In this paper, we show that variability used...
We present an adaptive control approach combining forward kinematics model learning methods with the operational space approach. This combination endows robot ability to realize hierarchically organised learned tasks in parallel, using null projectors built upon models. illustrate proposed method on a simulated 3 degrees of freedom planar robot. system is used as benchmark compare our alternative based extended Jacobian. show better versatility retained respect latter.
In this paper we address the problem of learning to recognize objects by manipulation in a developmental robotics scenario. life-long perspective, humanoid robot should be capable improving its knowledge with active perception. Our approach stems from cognitive development infants, exploiting curiosity-driven improve perceptual objects. These functionalities are implemented as perception, control and exploration modules part Cognitive Architecture MACSi project. integrate these into an...
Whole-body controllers provide the tools to execute multiple simultaneous tasks on humanoid robots, but given robot's internal and external constraints, interferences are often generated which impede task completion. Priorities can be assigned each manage these interferences, unfortunately, this is done at detriment of one or more tasks. In paper we present a novel framework for defining optimizing in order resolve potential prior execution remove need prioritization. Our parameterizes with...
Traditional industrial applications involve robots with limited mobility. Consequently, interaction (e.g. manipulation) was treated separately from whole-body posture balancing), assuming the robot firmly connected to ground. Foreseen augmented autonomy and physical Within this novel context, influences stability balance. To allow surpass barriers between control, forthcoming robotic research needs investigate principles governing motion coordination contact dynamics. There is a need of...
Many recent human-robot collaboration strategies, such as Assist-As-Needed (AAN), are promoting humancentered robot control, where the continuously adapts its assistance level based on real-time need of human counterpart.One fundamental assumptions these approaches is ability to measure or estimate physical capacity humans in real-time.In this work, we propose an algorithm for feasibility set analysis a generic class linear algebra problems.This novel iterative convex-hull method applied...
This paper proposes a robust whole-body control formulation for biped balance in disturbed conditions by manipulation tasks. In order to include the effects of interaction robot with its environment, required task control, we introduce distributed preview which captures both and behaviors enables regulation impedance. The initial ZMP is extended take into account disturbance resulting from adaptive impedances used drive upper-limbs. behavior illustrated simple scenario. Its aptitude...