Abstract
Cable-driven parallel robots have large workspaces compared to standard manipulator robots. In return, the presence of cables creates additionnal possible collisions, as well as positivity contraints on the cables tensions. The software platform HPP was developped by the GEPETTO team from the LAAS laboratory for motion planning and manipulation planning for articulated systems in cluttered environement.This thesis extends the algorithms within HPP to cable-driven parallel robots. The present research work provides solutions to the continuous trajectory validation for cable-driven parallel robots. First, a continuous collision detection algorithm is extended to take into account all the collisions pairs which include cables. This algorithm is based on the computation of an upper bound on the relative velocity between two bodies in order to succesively validate intervals of the trajectory. Secondly, this same continuous validation algorithm is extended to handle the validation of the cable tensions.Together, the continuous collision detection and the continous tension validation form a global continuous validation method for cable-driven parallel robots. This method is used within probabilistic planning algorithm to plan manipulation movements for cable robots. Several scenarios representing industrial tasks are tested with the robot CoGiRo both in simulation as well as on the real robot.