Summary: | Robotic arms are essential in a variety of industrial processes. However, the dexterous
workspace of a robotic arm is limited. This limitation can be overcome by making the robotic
arm mobile. Such robots, which comprise a robotic manipulator installed on a wheeled mobile
platform, are called mobile robots. A mobile manipulator can attain a position in space which
a robotic arm with fixed base may not be able to reach otherwise. To be applicable to a
variety of scenarios, these robots need to meet user-defined margins on their trajectory
tracking error, irrespective of the payload transported, faults, and failures. In this thesis,
we study the dynamics of mobile manipulator comprising both a differential-drive mobile
robot (DDMR) and a robotic arm. Thus, we design a model reference adaptive controller
(MRAC) for this mobile manipulator to regulate this vehicle and guarantee robustness to
uncertainties in the robot's inertial properties such as the mass of the payload transported
and friction coefficients. === Master of Science === Humans are able to perform tasks effectively owing to their extraordinary sense of perception and due to their ability to easily grasp things. Although humans are well-suited to
perform any process, within an industrial context, a variety of tasks might pose danger to
humans, like dealing with hazardous materials or working in extreme environments. Moreover, humans may suffer from fatigue while performing repetitive tasks. These considerations
gave rise to the idea of robots which could do the work for humans and instead of humans.
Mobile manipulators are a kind of robot that is well-suited for performing a variety of tasks
such as collecting, manipulating, and deploying objects from multiple locations. In order to
make robots perform a user-specified task, we need to study how the robot reacts to external
forces. This knowledge helps us derive a mathematical model for the robotic system. This
dynamical model would then be essential in controlling the motion of the robot. In this thesis, we study the dynamics of a mobile manipulator, which comprises a two-wheeled ground
platform and a five degrees-of-freedom robotic arm. The dynamical model of this mobile
robot is then employed to design a controller that guarantees user-defined margins of error
despite uncertainties in some properties, such as the mass of the payload transported.
|