208documents trouvés
Experiments on whole-body manipulation and locomotion with footstep real-time optimization
N.DANG, J.P.LAUMOND, F.LAMIRAUX
GEPETTO
Manifestation avec acte : IEEE-RAS International Conference on Humanoid Robots ( HUMANOIDS ) 2012 du 29 novembre au 01 décembre 2012, Osaka (Japon), 2012, 6p. , N° 12484
Lien : http://hal.archives-ouvertes.fr/hal-00727600
Diffusable
Plus d'informations
Abstract
This paper focuses on the experiments on the HRP-2 humanoid robot using a framework of manipulation and locomotion with real-time footstep adaptation. Two classes of experiments are presented. On the one hand, a grasping task at various height level illustrates a whole-body task in combination with locomotion. On the other, stepping over obstacle experiments illustrate the particularity of humanoid robots. In all presented examples, footsteps are considered as a part of the robot's kinematic chain and are resolved as an optimization problem along with other degrees of freedom of the robot. The environment is perceived by the stereo vision system mounted on the robot which closes the loop with the control through a online footstep adaptation scheme.
Reverse control for humanoid robot task recognition
S.HAK, N.MANSARD, O.STASSE, J.P.LAUMOND
GEPETTO
Revue Scientifique : IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, Vol.42, N°6, pp.1524-1537, Décembre 2012, DOI 10.1109/TSMCB.2012.2193614 , N° 12257
Lien : http://hal.archives-ouvertes.fr/hal-00697272
Diffusable
Plus d'informations
Abstract
Efficient methods to perform motion recognition have been developed using statistical tools. Those methods rely on primitives learning in a suitable space, for example, the latent space of the joint-angle and/or adequate task spaces. Learned primitives are often sequential : a motion is segmented according to the time axis. When working with a humanoid robot, a motion can be decomposed into parallel sub-tasks. For example, in a waiter scenario, the robot has to keep some plates horizontal with one of its arms, while placing a plate on the table with its free hand. Recognition can thus not be limited to one task per consecutive segment of time. The method presented in this paper takes advantage of the knowledge of what tasks the robot is able to do and how the motion is generated from this set of known controllers, to perform a reverse engineering of an observed motion. This analysis is intended to recognize parallel tasks that have been used to generate a motion. The method relies on the task-function formalism and the projection operation into the null space of a task to decouple the controllers. The approach is successfully applied on a real robot to disambiguate motion in different scenarios where two motions look similar but have different purposes.
Walking to grasp: modeling of human movements as invariants and an application to humanoid robotics
M.SREENIVASA, P.SOUERES, J.P.LAUMOND
University of Tokyo, GEPETTO
Rapport LAAS N°12111, DOI 10.1109/TSMCA.2011.2178830 , Mars 2012, 16p.
Diffusable
126765La robotique: une récidive d'Héphaïstos
J.P.LAUMOND
GEPETTO
Ouvrage (auteur) : Fayard, Les leçons inaugurales du Collège de France, N°ISBN 978-2-213-66905-2, Février 2012, 78p. , N° 12277
Non disponible
127379Dynamic walking makes humanoid robot small-space controllable: application to motion planning
S.DALIBARD, A.EL KHOURY, F.LAMIRAUX, M.TAIX, J.P.LAUMOND
GEPETTO
Rapport LAAS N°11697, Décembre 2011, 21p.
Lien : http://hal.archives-ouvertes.fr/hal-00654175
Diffusable
Plus d'informations
Abstract
This paper presents a general method for planning collision-free whole-body walking motions for humanoid robots. First, we present a randomized algorithm for constrained motion planning, that is used to generate collision-free statically balanced paths solving manipulation tasks. Then, we show that dynamic walking makes humanoid robots small-space controllable. Such a property allows to easily transform collision-free statically balanced paths into collision-free dynamically balanced trajectories. It leads to a sound algorithm which has been applied and evaluated on several problems where whole-body planning and walk are needed, and the results have been validated on a real HRP-2 robot.
Tracking an omnidirectional evader with a differential drive robot
R.MURRIETA-CID, U.RUIZ, J.L.MARROQUIN, J.P.LAUMOND, S.HUTCHINSON
CIMAT, GEPETTO, Illinois
Revue Scientifique : Autonomous Robots, Vol.31, N°4, pp.345-366, Novembre 2011 , N° 11582
Diffusable
125778A framework for manipulation and locomotion with realtime footstep replanning
N.DANG, F.LAMIRAUX, J.P.LAUMOND
GEPETTO
Manifestation avec acte : IEEE-RAS International Conference on Humanoid Robots (HUMANOIDS 2011), Bled (Slovénie), 26-28 Octobre 2011, pp.676-681 , N° 11846
Lien : http://hal.archives-ouvertes.fr/hal-00697564
Diffusable
Plus d'informations
Abstract
This paper focuses on realization of tasks with locomotion on humanoid robots. Locomotion and whole body movement are resolved as one unique problem. The same planner and controller are used for both stages of the movement. Final posture and footprint placements are found by resolving an optimization problem on the robot augmented by its footprints. Footstep replanning is done in realtime to correct perception and execution errors. The framework is demonstrated with the HRP-2 robot in a number of different scenarios
Small-space controllability of a walking humanoid robot
S.DALIBARD, A.EL KHOURY, F.LAMIRAUX, M.TAIX, J.P.LAUMOND
GEPETTO
Manifestation avec acte : IEEE-RAS International Conference on Humanoid Robots (HUMANOIDS 2011), Bled (Slovénie) , 26-28 Octobre 2011, pp.739-744 , N° 11367
Lien : http://hal.archives-ouvertes.fr/hal-00602384/fr/
Diffusable
Plus d'informations
Abstract
This paper presents a two-stage motion planner for walking humanoid robots. A first draft path is computed using random motion planning techniques that ensure collision avoidance. In a second step, the draft path is approximated by a whole-body dynamically stable walk trajectory. The contributions of this work are: (i) a formal guarantee, based on smalltime controllability criteria, that the first draft path can be approximated by a collision-free dynamically stable trajectory; (ii) an algorithm that uses this theoretical property to find a solution trajectory. We have applied our method on several problems where whole-body planning and walk are needed, and the results have been validated on a real platform: the robot HRP-2.
Planning foot placements for a humanoid robot: a problem of inverse kinematics
O.KANOUN, J.P.LAUMOND, E.YOSHIDA
GEPETTO, AIST
Revue Scientifique : International Journal of Robotics Research, Vol.30, N°4, pp.476-485, Avril 2011, doi:10.1177/0278364910371238 , N° 10616
Diffusable
122818Linear dimensionality reduction in random motion planning
S.DALIBARD, J.P.LAUMOND
GEPETTO
Rapport LAAS N°11079, doi: 10.1177/0278364911403335, Février 2011, 29p.
Lien : http://hal.archives-ouvertes.fr/hal-00486793/fr/
Diffusable
Plus d'informations
Abstract
The paper presents a method to control probabilistic diffusion in motion planning algorithms. The principle of the method is to use on line the results of a diffusion algorithm to describe the free space in which the planning takes place, by computing a Principal Component Analysis (PCA). This method identifies the locally free directions of the free space. Given that description, our algorithm accelerates the diffusion along these favoured directions. That way, if the free space appears as a small volume around a submanifold of a highly dimensioned configuration space, the method overcomes the usual limitations of diffusion algorithms and finds a solution quickly. The presented method is theoretically analyzed and experimentally compared to known motion planning algorithms.