Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Yıldız Teknik Üniversitesi, Elektrik-Elektronik Fakültesi, Kontrol Ve Otomasyon Mühendisliği, Türkiye
Tezin Onay Tarihi: 2021
Tezin Dili: İngilizce
Öğrenci: Özgür Acar
Danışman: Claudia Fernanda Yaşar
Özet:
Wall-climbing
robots are generally used for dangerous and expensive tasks when performed by
humans due to severe conditions. The design, modelling, and control of a
wall-climbing robot for tank inspection are presented in this thesis. A novel
mechanical design for the system actuation based on a palletized robot with magnetic
adhesion is proposed. The magnetic adhesion system is used to attach the robot
to a metallic wall. The locomotion of climbing robots still represents
challenges when designing control systems for each particular adhesion
mechanism. The main goal is to build a light robot that includes all the
required hardware, mechanism, actuators, and sensors that will allow for the
use of reliable control systems and inspection tasks.
The
sensor and acquisition system for robot navigation is explained in this thesis.
A sensory-data strategy with a sensor fusion is used to estimate the actual
position of the robot. To track the required trajectories sensor measurements
and the robot kinematics are used, along with a Kalman filter, which, in turn,
would enhance the motion control strategy, making the system robust to the
possible disturbances and noise. A camera-based inspection system is part of
the sensory system but is used for inspection purposes.
The
robot system was designed with the capability of climbing metallic vertical
sheets. Experiments confirmed a successful performance of system mechanics,
sensory fusion strategy, and the control performance for robot locomotion and
adhesion. This climbing robot has demonstrated the capability to follow an
accurate reference on a vertical wall. The robotic application is endowed with
a camera system that can detect, process, classify corrosion figures, and
present classification results on a monitory diagnostic system.
Finally,
to fulfil the high robustness requirements, a control method is based on the
design of a PD controller and an observer, in which the overall disturbance,
Coulomb friction, and sudden changes of the inertia, are estimated and then
compensated. A robust motion control strategy for the climbing robot mobility
that uses DC motors is explained. This method is part of a preliminary study to
design advanced motion controllers for a climbing strategy. The controller is
still under development, but some designs from the availed literature and
initial simulations have been performed with successful results. Simulations of the control strategy were
carried out using Matlab/Simulink. Further work aims at validating advanced
control methods for climbing robots under high disturbances coming from
magnetic, Coulomb, and gravity forces.