Motion planning of a mobile manipulator using fuzzy controller to dexterity measures

Abstract

This paper proposes a new approach for motion planning of a redundant mobile manipulator. The motion planning of mobile manipulator addresses the problem of trajectory following by the end-effector of manipulator in coordination with the mobile platform. In the proposed approach, a controller has been designed based on Takagi-Sugeno fuzzy system and trained to find the appropriate position of the mobile robot while following the given trajectory by maximizing the manipulability measure of the manipulator. The movement of mobile robot in coordination with the manipulator is depend on the manipulability measure of the manipulator. Moreover, the motion of the end-effector should be cognizance by the mobile base, if the task-space is outside the good manipulability boundary, while the motion of the manipulator should be cognizance by the manipulator, if the task-space is inside the good manipulability boundary. Therefore, the decomposition of the motion of the manipulator and motion of the mobile platform has to be done based on the task-space location of the end-effector. The proposed methodology is proved to be a very efficient in terms of computational cost to the fuzzy controller, which can be implemented in real-time.