Planification de trajectoire d'un manipulateur a l'aide des reseaux neuroniques et de la logique floue: Une application a la saisie d'un objet mobile

Abstract

Ce memoire presente une etude du potentiel des reseaux de neurones et de la logique floue en tant que methodes alternatives pour la planification de trajectoire d'un manipulateur seriel plan a trois degres de liberte. Une structure de reseaux neuroniques est proposee pour l'anticipation de la position, de l'orientation, de la vitesse et de l'acceleration d'un objet en mouvement. Un algorithme de planification de trajectoire a base de logique floue traitant aussi l'evitement de collisions entre l'objet et les membres du robot est decrit en details. Les concepts generaux servant a la determination d'un point de saisie, a l'optimisation des modules, ainsi qu'a la gestion de l'approche de l'objet par l'organe terminal sont presentes. Enfin, les resultats obtenus en simulation sont illustres et analyses afin de conclure sur la robustesse et le potentiel important des reseaux neuroniques et de la logique floue dans le developpement d'algorithmes de commande de haut niveau appliques a la robotique.

Alternate abstract:

You are viewing a machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer

Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer

This thesis presents a study of the potential of neural networks and fuzzy logic as alternative methods for the trajectory planning of a planar serial manipulator with three degrees of freedom. A structure of neural networks is proposed for the anticipation of the position, the orientation, the speed and the acceleration of a moving object. A fuzzy logic-based trajectory planning algorithm also dealing with the avoidance of collisions between the object and the robot's limbs is described in detail. The general concepts used for the determination of a point of seizure, the optimization of the modules, as well as the management of the approach of the object by the terminal organ are presented. Finally, the results obtained in simulation are illustrated and analyzed in order to conclude on the robustness and the significant potential of neural networks and fuzzy logic in the development of high-level control algorithms applied to robotics.