國立高雄大學圖資館 |

Language: English

Back

整合區間型卡爾曼濾波器與模糊控制系統於智慧型機器人之自主式避障與導航 =...

國立高雄大學電機工程學系碩士班

整合區間型卡爾曼濾波器與模糊控制系統於智慧型機器人之自主式避障與導航 = Integrating interval Kalman filtering and fuzzy sets for obstacle avoidance and navigation of mobile robots

Record Type:	Language materials, printed : monographic
Paralel Title:	Integrating interval Kalman filtering and fuzzy sets for obstacle avoidance and navigation of mobile robots
Author:	黃右棟,
Secondary Intellectual Responsibility:	國立高雄大學
Place of Publication:	[高雄市]
Published:	撰者;
Year of Publication:	2013[民102]
Description:	81面圖，表 : 30公分;
Subject:	超音波
Subject:	sonar
Online resource:	http://handle.ncl.edu.tw/11296/ndltd/15158597192591958004
Notes:	105年10月25日公開
Notes:	參考書目：面68-71
Summary:	各式感應器的量測結果都存在不確定性與誤差，為取得較精確合理的數據，誤差修正是其重要的一環。卡爾曼濾波器 (Kalman filtering) 是常用來作為修正量測誤差的方法之一，但傳統卡爾曼濾波器對訊號急遽變化或存在不確定性時，其修正精確度往往有所限制。區間型卡爾曼濾波器 (interval Kalman filtering) 為卡爾曼濾波器的延伸，可以處理訊號的不確定性，但訊號因為區間計算結果而累積，使得在實用上發生困難。為有效解決其此限制，使區間型卡爾曼濾波器能有效應用於智慧型機器人的導航與避障，本研究利用區間型卡爾曼濾波器取得感測器預測之區間值，提出應用權重概念對預測值作相對修正之方法，以獲得較精確的預測效果，提升其實用性。該方法稱為區間型權重卡爾曼濾波器 (weighted interval Kalman filtering)。本研究利用區間型權重卡爾曼濾波結合模糊控制 (fuzzy logic control)，在裝載多組超音波感應器模組之四輪型智慧型機器人進行避障。超音波感應器訊號，經區間型權重卡爾曼濾法過濾後，作為模糊控制輸入值，進行機器人自主式動態避障導航之任務。本方法經各式環境測試顯示，本研究提出之區間型權重卡爾曼濾波法的效果在環境急遽變化大時，訊號處理效果優於傳統卡爾曼濾波。 Due to the inaccuracy and imperfection of sensors, signals from sensors need to be filtered out before they are used for advanced processing. Kalman filtering (KF) is such a technique being able to filter out temporary signal disturbance. The interval Kalman filtering (IKF) is an advanced version of KF, in which all signals are wrapped as numerical intervals and outperforms traditional KF. For practical uses, an output estimation from KF or IKF is a certain value. The thesis presents a weighting heuristics that consider the strength of positive/negative intervals of the signal estimations and produce a weighted output to be used as input to fuzzy controllers of a mobile robot. The method is named as the weighted interval Kalman filtering (WIKF). The study integrates WIKF and fuzzy logic control (FLC) on a four-wheeled mobile robot for obstacle avoidance and autonomous navigation. The mobile robot used in this study carries 9 simple sonar sensors that acquire signals of obstacles periodically. Input to WIKF are sensor measurements each of which is with an interval describing the range of measurement errors; the output of WIKF is an interval of estimate of sensor signal. For robot controls, a certain value is calculated from the interval output of WIKF. In the thesis, several experiments of obstacle avoidance are conducted on a mobile robot and the results are analyzed. The results show that WIKF provides reliable navigation information better than that from traditional KF and IKF.

整合區間型卡爾曼濾波器與模糊控制系統於智慧型機器人之自主式避障與導航 = Integrating interval Kalman filtering and fuzzy sets for obstacle avoidance and navigation of mobile robots
黃, 右棟

整合區間型卡爾曼濾波器與模糊控制系統於智慧型機器人之自主式避障與導航 = Integrating interval Kalman filtering and fuzzy sets for obstacle avoidance and navigation of mobile robots / 黃右棟撰 - [高雄市] : 撰者, 2013[民102]. - 81面 ; 圖，表 ; 30公分.
105年10月25日公開參考書目：面68-71.
超音波sonar

整合區間型卡爾曼濾波器與模糊控制系統於智慧型機器人之自主式避障與導航 = Integrating interval Kalman filtering and fuzzy sets for obstacle avoidance and navigation of mobile robots
LDR:04318nam0 2200277 450 001 389736
005 20161117141435.0
010 0 $b 精裝
010 0 $b 平裝
100 $a 20130926d2013 k y0chiy50 e
101 1 $a chi $d chi $d eng
102 $a tw
105 $a ak am 000yy
200 1 $a 整合區間型卡爾曼濾波器與模糊控制系統於智慧型機器人之自主式避障與導航 $d Integrating interval Kalman filtering and fuzzy sets for obstacle avoidance and navigation of mobile robots $z eng $f 黃右棟撰
210 $a [高雄市] $c 撰者 $d 2013[民102]
215 0 $a 81面 $c 圖，表 $d 30公分
300 $a 105年10月25日公開
300 $a 參考書目：面68-71
314 $a 指導教授：吳志宏博士
328 $a 碩士論文--國立高雄大學電機工程學系碩士班
330 $a 各式感應器的量測結果都存在不確定性與誤差，為取得較精確合理的數據，誤差修正是其重要的一環。卡爾曼濾波器 (Kalman filtering) 是常用來作為修正量測誤差的方法之一，但傳統卡爾曼濾波器對訊號急遽變化或存在不確定性時，其修正精確度往往有所限制。區間型卡爾曼濾波器 (interval Kalman filtering) 為卡爾曼濾波器的延伸，可以處理訊號的不確定性，但訊號因為區間計算結果而累積，使得在實用上發生困難。為有效解決其此限制，使區間型卡爾曼濾波器能有效應用於智慧型機器人的導航與避障，本研究利用區間型卡爾曼濾波器取得感測器預測之區間值，提出應用權重概念對預測值作相對修正之方法，以獲得較精確的預測效果，提升其實用性。該方法稱為區間型權重卡爾曼濾波器 (weighted interval Kalman filtering)。本研究利用區間型權重卡爾曼濾波結合模糊控制 (fuzzy logic control)，在裝載多組超音波感應器模組之四輪型智慧型機器人進行避障。超音波感應器訊號，經區間型權重卡爾曼濾法過濾後，作為模糊控制輸入值，進行機器人自主式動態避障導航之任務。本方法經各式環境測試顯示，本研究提出之區間型權重卡爾曼濾波法的效果在環境急遽變化大時，訊號處理效果優於傳統卡爾曼濾波。 Due to the inaccuracy and imperfection of sensors, signals from sensors need to be filtered out before they are used for advanced processing. Kalman filtering (KF) is such a technique being able to filter out temporary signal disturbance. The interval Kalman filtering (IKF) is an advanced version of KF, in which all signals are wrapped as numerical intervals and outperforms traditional KF. For practical uses, an output estimation from KF or IKF is a certain value. The thesis presents a weighting heuristics that consider the strength of positive/negative intervals of the signal estimations and produce a weighted output to be used as input to fuzzy controllers of a mobile robot. The method is named as the weighted interval Kalman filtering (WIKF). The study integrates WIKF and fuzzy logic control (FLC) on a four-wheeled mobile robot for obstacle avoidance and autonomous navigation. The mobile robot used in this study carries 9 simple sonar sensors that acquire signals of obstacles periodically. Input to WIKF are sensor measurements each of which is with an interval describing the range of measurement errors; the output of WIKF is an interval of estimate of sensor signal. For robot controls, a certain value is calculated from the interval output of WIKF. In the thesis, several experiments of obstacle avoidance are conducted on a mobile robot and the results are analyzed. The results show that WIKF provides reliable navigation information better than that from traditional KF and IKF.
510 1 $a Integrating interval Kalman filtering and fuzzy sets for obstacle avoidance and navigation of mobile robots $z eng
610 # 0 $a 超音波 $a 卡爾曼濾波器 $a 區間型卡爾曼濾波器 $a 模糊控制 $a 避障 $a 導航 $a 自主式機器人
610 # 1 $a sonar $a Kalman filtering $a interval Kalman filtering $a fuzzy controller $a obstacle avoidance $a navigation $a autonomous mobile robots
681 $a 008M/0019 $b 542201 4444 $v 2007年版
700 1 $a 黃 $b 右棟 $4 撰 $3 614593
712 0 2 $a 國立高雄大學 $b 電機工程學系碩士班 $3 166118
801 0 $a tw $b NUK $c 20161014 $g CCR
856 7 # $u http://handle.ncl.edu.tw/11296/ndltd/15158597192591958004 $2 http $z PDF全文