hal-00484853, version 1
Fast Resolution of Hierarchized Inverse Kinematics with Inequality Constraints
Adrien Escande 1Nicolas Mansard
2Pierre-Brice Wieber 3, 4
IEEE International Conference on Robotics and Automation (2010) 7p.
Résumé : Classically, the inverse kinematics is performed by computing the singular value decomposition of the matrix to invert. This enables a very simple writing of the algorithm. However, the computation cost is high, especially when applied to complex robots and complex sets of constraints (typically around 5ms for 50 degrees of freedom -- DOF). In this paper, we propose a dedicated adaptation of quadratic programming that enables fast computations of the hierarchical inverse kinematics (around 0.1ms for 50~DOF). We then extend this algorithm to deal with unilateral constraints, obtaining sufficiently high performances for reactive control.
- 1 : Laboratoire d'Intégration des Systèmes et des Technologies (CEA LIST)
- CEA : DRT/LIST
- 2 : Laboratoire d'analyse et d'architecture des systèmes (LAAS)
- CNRS : UPR8001 – Université Paul Sabatier - Toulouse III – Institut National Polytechnique de Toulouse - INPT – Institut National des Sciences Appliquées de Toulouse
- 3 : Joint Robotics Laboratory AIST & CNRS (JRL)
- AIST – CNRS : UMI3218
- 4 : BIPOP (INRIA Grenoble Rhône-Alpes / LJK Laboratoire Jean Kuntzmann)
- INRIA – Laboratoire Jean Kuntzmann
- Domaine : Informatique/Robotique
- Mots-clés : hierarchy of tasks – kinematics – redundancy – quadratic problem
- Référence interne : Rapport LAAS n° 10175
- hal-00484853, version 1
- http://hal.archives-ouvertes.fr/hal-00484853
- oai:hal.archives-ouvertes.fr:hal-00484853
- Contributeur : Nicolas Mansard
- Soumis le : Mercredi 19 Mai 2010, 12:41:33
- Dernière modification le : Lundi 21 Février 2011, 16:41:43
Documents associés
Exporter