الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 78 |  |  |
| | | from | 78 | | |
Abstract
A new wave of technology powered by mobile manipulators enables us to
be versatile in many filed, such as construction, space, production lines, and
warehouses. The main objective of the present work is to study the kinematics of a
6 DOF arm manipulator mounted on a high maneuverability mecanum wheeled
mobile base and solve the inverse kinematics of the arm using the Neural Networks
technique.
The current thesis investigates a mobile manipulator with six degrees of
freedom (DOFs) that can track a desired trajectory. A motion planning approach is
proposed based on examining its kinematics models. The forward and inverse
kinematics are analyzed using the Denavit-Hartenberg technique. The end-location
effector and orientation are divided into two sections. In the first part, the
manipulator gives sub-vectors like position and orientation projected on the Z-axis
in the world frame. The mobile base and manipulator follow the necessary path in
the second half to get to the sub-vectors on the X and Y axes of the world frame,
respectively. The efficacy of the suggested technique is demonstrated using
simulated outcomes.
The curves and results of the extracted experiment of the mobile base support
the accuracy, viability, and excellent maneuverability. The arm manipulator’s
kinematic redundancy is considered while solving the inverse issue.
The Neural network model is used to address the inverse kinematics problem,
made non-linear solving systems easier than traditional techniques for solving such
complicated systems. An intermediary step before gathering the training data of
117,000 randomly selected points from the Mobile Manipulator’s entire workspace
employed at the neural network was to generate 6 million points using an
implemented method.