Fuzzy Ship Autopilot Design based on Left-right and Rear Leading Mark Concept

碩士 === 國立臺灣海洋大學 === 通訊與導航工程學系 === 101 === This work aims at studying the performance of fuzzy autopilots of different number of inputs while using the leading mark based visual guidance for accomplishing small–boat approaching maneuvers. The experiment uses a small FRP boat as the test platform and...

Full description

Bibliographic Details
Main Author: 李康維
Other Authors: Ching-Yaw Tzeng
Format: Others
Language:zh-TW
Published: 2013
Online Access:http://ndltd.ncl.edu.tw/handle/96553265409195594944
Description
Summary:碩士 === 國立臺灣海洋大學 === 通訊與導航工程學系 === 101 === This work aims at studying the performance of fuzzy autopilots of different number of inputs while using the leading mark based visual guidance for accomplishing small–boat approaching maneuvers. The experiment uses a small FRP boat as the test platform and the visual guidance schemes adopt the left-right and rear leading mark concept and a fuzzy logic autopilot is employed to guide the vessel toward the intended water way. In the proposed approach, three canvas targets are arranged at left-right and rear position as leading marks on the quayside. These targets are detected by a Charge Coupled Device (CCD) camera and the resulting image is processed using a color recognition scheme to accurately detect the location of the targets in the image by using LabVIEW/Vision Builder AI software. The geometric centers of these targets are then calculated, and the separation distance between them are used to compute the deviated heading angle and the deviated position angle needed as feedback error signals in the adopted autopilots. The autopilot controller design process can be divided into three parts, the first part uses the deviated heading angle (θLM), and the ship's deviated position angle (θCR) as two fuzzy inputs. The second part adopts the deviated heading angle (θLM), the ship's deviated position angle (θCR) and the ship's position skew angle (θVCR) as three fuzzy inputs. With three fuzzy inputs, a total of twenty-seven fuzzy rules are generated. A method to simplify the twenty-seven fuzzy rules was proposed in this study. The third part of the controller design method is an enhancement of part two. Namely, to combine the deviated heading angle (θLM), and the ship's deviated position angle (θCR) into one input and the ship's position skew angle (θVCR) as another input, and a refined two inputs fuzzy logic controller is proposed. Two kinds of comparisons are made in the experimental studies. First one was based on the three different types of input controller experiment comparison studies. The second focused on the comparison of original three-input controller fuzzy rules experiments and the simplified three-input controller fuzzy rules. The feasibility of the proposed concept was verified by performing a series of FPR boat based experiments conducted in the waterways of the NTOU small boat harbor. The experimental results demonstrate that the proposed fuzzy rules simplification design is valid, and the left-right and rear leading mark design performs better than the original left-right mark design, and the proposed enhanced guidance scheme also out performs the original left-right mark design.