Design and modelling of an autonomous robotic airship with soft computing control

Wong, Wei Kitt (2012) Design and modelling of an autonomous robotic airship with soft computing control. Masters thesis, Universiti Malaysia Sabah.

[img]
Preview
Text
mt0000000639.pdf

Download (10MB) | Preview

Abstract

UAV (Unmanned Aerial Vehicle) can be defined as an autonomously / remotely piloted air vehicle. UAV development has seen a constant rise due to the developement of highly reliable sensors and the Global Positioning System (GPS). UAVs control system remains dominated by model based control system that can be unreliable in the absence of highly precise instruments and non linearities. In recent years, with the development of advance soft computing control methods, countless research work have applied these controls system In their autonomous systems. The objective of this thesis is to apply the concept of Artificial Inteligence in UAV developement To meet this objective, a prototype Airship is designed with on board sensors and GPS receivers. In this dissertation, fuzzy logic was proposed for the autonomous airship.The navigation technique proposed is the waypoint following algorithm where a path is separated Into a series of waypoints. The fuzzy controller considers the differrence between the calculated bearing from the current position and the target position. It then gives an apropriate propulsion value to the actuators. The actuators use the pulse width modulation as the signals. A servo controller was developed to control the respective actuators, the motor speed controllers and the vectoring thrust servo. The PWM based actuator controller was constructed by varying the delay between the "on" and "off" time of pin connected to the actuators. An on-board microcontroller as used to stream the GPS and other telemetry data to the main processor wirelessly. The GPS and telemetry data were processed using string recognition methods.The processed and filtered telemetry and GPS data were fed to the fuzzy logic system. The proposed control system was tested on a simulation level before flight tested on a prototype airship. The control system were tested on simulation which were developed from simulink. The result shows a feasible means of UAV control. Along with the prototype airship, a path planner was also proposed for the UAV for the purpose of obstacle avoidance and path optimization using genetic algorithm to minimize on the distance travelled and to generate the shortest path to cover the specified waypoints. The concept path planner was demonstrated on a cartesian grid.

Item Type: Thesis (Masters)
Keyword: Unmanned Aerial Vehicle (UAV), Global Positioning System (GPS), control system, Artificial Intelligence, Airship, cartesian grid, robotic
Subjects: T Technology > TJ Mechanical engineering and machinery
Department: SCHOOL > School of Engineering and Information Technology
Depositing User: ADMIN ADMIN
Date Deposited: 01 Sep 2015 15:38
Last Modified: 07 Nov 2017 15:01
URI: https://eprints.ums.edu.my/id/eprint/11581

Actions (login required)

View Item View Item