54documents trouvés
Interactive motion planning for assembly tasks
D.FLAVIGNE, M.TAIX, E.FERRE
GEPETTO, Kineo
Manifestation avec acte : International Symposium on Robot and Human Interactive Communication, RO-MAN 2009, Toyama (Japon), 27 Septembre-2 Octobre 2009, pp.430-435 , N° 09131
Diffusable
119280Visibility local tree for motion planning
D.FLAVIGNE, M.TAIX
GEPETTO
Rapport LAAS N°09355, Juin 2009, 6p.
Diffusable
118095Haptic assembly and disassembly task assistance using interactive path planning
N.LADEVEZE, J.Y.FOURQUET, B.PUEL, M.TAIX
ENIT Tarbes, ALSTOM Transport, GEPETTO
Manifestation avec acte : IEEE Virtual Reality Conference (VR 2009), Lafayette (USA), 14-18 Mars 2009, 7p. , N° 09016
Diffusable
Plus d'informations
Mots-Clés / Keywords
Virtual reality; Motion planning; Octree; A star; Rapidly exploring deterministic tree; Haptic;
Interactive motion planning for assembly tasks
D.FLAVIGNE, M.TAIX
GEPETTO
Rapport LAAS N°08598, Décembre 2008, 6p.
Diffusable
115645A principled approach to biological motor control for generating humanoid robot reaching movements
T.M.TUAN, P.SOUERES, M.TAIX, E.GUIGON
GEPETTO, INSERM ANIM
Manifestation avec acte : IEEE International Conference on Biomedical Robotics and Biomechatronics, Scottsdale (USA), 19-22 Octobre 2008, pp.783-788 , N° 08218
Diffusable
Plus d'informations
Abstract
This paper presents an application of neurobiological motor principles to the control of humanoid robot reaching movements. The model is mainly based on the hypotheses that the energy of motoneurons signals is continuously minimized along the motion and that dynamic and static efforts are processed separately. This paradigm is used to control the robot HRP2 by considering a dynamic model of the arm including, for each degree of freedom, two second order low-pass filters modeling the neuromuscular system defined by a pair of antagonist muscles. The optimal control problem is solved as a nonlinear programming problem by using a direct transcription method coupled with the optimization software Ipopt. This approach allows to generate realistic movements with anthropomorphic features such as quasi-rectilinear trajectories and bell-shaped velocity.
Mots-Clés / Keywords
Neurobiological principles; Reaching control; Humanoid robot ; Direct transcription; Human motor control; Nonlinear programming;
Interactive motion planning in virtual reality environments
N.LADEVEZE, J.Y.FOURQUET, M.TAIX
GEPETTO, ENIT Tarbes
Manifestation avec acte : 10th Virtual Reality International Conference, Laval (France), 9-11 Avril 2008, 7p. , N° 08121
Diffusable
Plus d'informations
Mots-Clés / Keywords
Virtual reality; Motion planning; Unbalanced octree; A*; Rapidly-exploring random tree; Interactive motion planning; Haptic; Force feedback;
Planning robust landmarks for sensor based motion
M.TAIX, A.C.MALTI, F.LAMIRAUX
GEPETTO
Manifestation avec acte : European Robotics Symposium (EUROS 2008), Prague (République Tchèque), 26-28 Mars 2008, 7p. , N° 07152
Diffusable
Plus d'informations
Abstract
Our work is focused on defining a generic approach for planning landmark based motion. In previous works we have introduced the formal basis for this and showed simulation results. In this paper we first demonstrate the relevance of our work with experiments on a real robot. Then, on the base of these results we introduce new strategy for planning selecting landmarks in order to improve the robustness of the navigation task in a cluttered environment.
A paradigm of neurobiological motor principles for the control of humanoid robots reaching motions
T.M.TUAN, P.SOUERES, M.TAIX, E.GUIGON
GEPETTO, INSERM ANIM
Manifestation avec acte : ESF-JSPS Frontier Science Conference for Young Researchers Experimental Cognitive Robotics, Kanagawa (Japon), 9-15 Mars 2008, 16p. , N° 07695
Diffusable
113892Sensor based planning: experimental results
M.TAIX, A.C.MALTI, F.LAMIRAUX
GEPETTO
Rapport LAAS N°07215, Mai 2007, 7p.
Diffusable
Plus d'informations
Mots-Clés / Keywords
Sensor based motion; Mobile robot; Sensor planning; Motion planning;
Toward robust landmark selection for landmark based motion
A.C.MALTI, M.TAIX, F.LAMIRAUX
RIA, RIS, GEPETTO
Rapport LAAS N°06589, Septembre 2006, 8p.
Diffusable
107795