English  |  正體中文  |  简体中文  |  Items with full text/Total items : 78852/78852 (100%)
Visitors : 35385936      Online Users : 1019
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version

    Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/93327

    Title: 以 AI 與 PSD 輔助開發先進人行環境多功能檢測儀之研究;Study of Developing Advanced Pedestrian Environment Multifunctional Detection Instrument with AI and PSD Assistance
    Authors: 黃振愷;Huang, Cheng-Kai
    Contributors: 土木工程學系
    Keywords: 人行道檢測儀;人行道指標評估;功率譜密度;Sidewalk survey instrument;Sidewalk Index Evaluation;Power Spectral Density.
    Date: 2024-01-29
    Issue Date: 2024-03-05 16:23:03 (UTC+8)
    Publisher: 國立中央大學
    Abstract: 臺灣內政部國土管理署每年邀請專家學者對當地政府所管養之人行環境進行檢查與評估,本研究在人行環境調查車的基礎上增加增進系統整合與效能、以陀螺儀取代加速度規、增加雷射,建立的先進人行環境多功能檢測儀在配合人工智慧預測模型(Artificial Intelligence Model, AI)及功率譜密度(Power Spectral Density, PSD)分析演算法輔助建立人行環境多功能檢測儀,可評估包含舒適性指標、人因危害指標、安全性指標以及暢行性指標,參考CNS 15731與ISO 13473-1規範進行驗證並使用相關濾波處理後可符ASTM E3028的要求量測輪椅平坦度指標(Wheelchair Pathway Roughness Index, WPRI),儀器可量測之紋理空間頻率範圍從0到1560 cycle/m,依照ISO 13473-1之分級可涵蓋Mega, Macro texture。
    本研究參考內政部國土管理署「112市區道路養護管理暨人行環境無障礙考評計畫」選定臺北市、新北市、桃園市、台中市、臺南市、高雄市廓進行人行道調查,舒適性指標結果依照分級表進行人行道WPRI等級分類,整體資料分別佔有very good:3.1%、good:67.6%、fair:27.2%、poor:1.8%、very poor:0.1%和impassable:0.2%;人因危害指標顯示全檢測段平均有10.5%的資料為暴露在大於3m/s2的高風險振動下,其可能會對輪椅使用者造成相關之危害風險;安全性指標參考相關文獻後使用快速傅立葉轉換(Fast Fourier Transform, FFT)和希爾伯特黃轉換(Hilbert Huang Transform, HHT)計算並選出特徵PSD參數,並輸入AI 基因編程演算法(Gene Expression Programming, GEP)建立現地測量剖面與英式擺錘數值(British Pendulum Number, BPN)的預測方程式,使用FFT建立的預測方程式具有高度解釋力且模型精度經平均絕對百分比誤差(Mean Absolute Percentage Error, MAPE)判定屬Good等級;暢行性指標的點雲合程結果正確勾畫出現檢測路徑,沿途各設施物可使用歐式距離分群正確辨識,人工淨寬量測與掃描之結果經MAPE值判定屬Excellent等級,代表此系統可替代人工量測的同時可進行全斷面掃描,大幅減輕現地人力。
    ;The Ministry of the Interior′s National Land Administration in Taiwan annually invites experts and scholars to inspect and assess the pedestrian environments managed by local governments. This study enhances the functionality of a pedestrian environment survey vehicle by integrating systems and improving efficiency. It replaces the accelerometer with a gyroscope, adds a 3D camera, and establishes an advanced multifunctional detection instrument for pedestrian environments. This instrument, assisted by an Artificial Intelligence Model (AI) and Power Spectral Density (PSD) analysis algorithms, evaluates comfort indicators, human hazard indicators, safety indicators, and mobility indicators. It adheres to CNS 15731 and ISO 13473-1 standards, and after relevant filtering processes, it meets the requirements for measuring the Wheelchair Pathway Roughness Index (WPRI) as per ASTM E3028. The instrument can measure texture spatial frequencies ranging from 0 to 1560 cycles/m, covering Mega and Macro textures according to ISO 13473-1 classifications.
    This study conducted pedestrian surveys in Taipei City, New Taipei City, Taoyuan City, Taichung City, Tainan City, and Kaohsiung City, referencing the Ministry of the Interior′s National Land Administration′s "112 Urban Road Maintenance Management and Pedestrian Environment Accessibility Assessment Plan." Comfort indicators classify WPRI levels according to a grading table, with the overall data distribution as follows: very good (3.1%), good (67.6%), fair (27.2%), poor (1.8%), very poor (0.1%), and impassable (0.2%). The human hazard indicator reveals that, on average, 10.5% of the data in the entire surveyed sections are exposed to vibrations exceeding 3 m/s², posing a high-risk hazard to wheelchair users. Safety indicators utilize Fourier and Hilbert-Huang transforms to calculate PSD parameters, inputting them into Gene Expression Programming (GEP) for establishing prediction equations for on-site measured profiles and British Pendulum Numbers (BPN). The developed equations exhibit high explanatory power (R²=0.7), MAPE (Mean Absolute Percentage Error) suggests the equation accuracy is Good. Mobility indicators show accurate delineation of the survey path, correct identification of facilities along the way using Euclidean distance clustering. The MAPE between manual width measurements and scanned results suggests the system accuracy is Excellent.
    The result indicates that the system, while replacing manual measurements, can perform full cross-sectional scans, significantly reducing on-site labor. The study successfully integrates and develops an evaluation system for comfort, human hazard, safety, and mobility indicators.
    Appears in Collections:[土木工程研究所] 博碩士論文

    Files in This Item:

    File Description SizeFormat

    All items in NCUIR are protected by copyright, with all rights reserved.

    社群 sharing

    ::: Copyright National Central University. | 國立中央大學圖書館版權所有 | 收藏本站 | 設為首頁 | 最佳瀏覽畫面: 1024*768 | 建站日期:8-24-2009 :::
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 隱私權政策聲明