以 AI 與 PSD 輔助開發先進人行環境多功能檢測儀之研究

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：135

、訪客IP：18.225.95.229

姓名

黃振愷(Cheng-Kai Huang) 查詢紙本館藏

畢業系所

土木工程學系

論文名稱

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)

相關論文

★ Engineering and Environmental Analysis of Maintenance Interval in Taiwan Freeway -the Case of Guanxi Section	★ 3D 鋪面調查車之驗證與國道應用分析
★ 營建剩餘土石方物流監控及管理系統之建置	★ 透水性鋪面保水與溫差成效之評估 -以中壢市龍慈路為例
★ 以生命週期評估滾筒碴與轉爐石應用於瀝青混凝土之研究	★ 傳統單點雷射與2D雷射應用於平坦度之比較研究
★ 鋪面劣化影像自動辨識應用於鋪面巡查精進研究	★ 自動化鋪面破壞影像辨識系統導入鋪面破壞維護管理系統之研究
★ 以ETC大數據結合FWD建立台灣區高速公路鋪面結構評估準則之研究	★ The Comparison Study of Various Surface Maintenance Alternatives in Taiwan Freeway
★ The Preliminary Study of conducting Pavement Maintenance Model for Taiwan Provincial Highways using Life-Cycle Cost Analysis	★ 冷拌再生瀝青混凝土應用於管線挖掘回填層之可行性研究
★ 台灣現行修補材料運用於柔性鋪面表層裂縫與坑洞修補之耐久性初步探討	★ 以車載藍光雷射建構國道鋪面抗滑值與二維紋理之關聯模型
★ 以不同光譜雷射應用於鋪面平坦度量測之綜合性評估	★ 冷拌再生瀝青混凝土應用於道路管線挖掘回填工程之現地驗證

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

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。
本研究所開發的先進人行環境多功能檢測儀同時可以符合四大指標的數據收集以簡化現地調查的繁雜，同時具有可連續並長時間檢測的能力，不同於人工調查只能進行點狀抽查，本研究可依照推行路線進行線性掃描，大幅提升時間效益，6都的大面積現地調查後提供臺灣人行道現況資料，幫助研究者及工程師瞭解臺灣人行道現況及未來的改進方向。

摘要(英)

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.

關鍵字(中)

★ 人行道檢測儀、人行道指標評估、功率譜密度

關鍵字(英)

★ Sidewalk survey instrument, Sidewalk Index Evaluation, Power Spectral Density.

論文目次

目錄 II
圖目錄 VI
表目錄 X
摘要 XII
Abstract XIII
一、緒論 1
1-1 研究背景 1
1-2 研究目的 1
1-3 研究範圍與限制 2
二、文獻回顧 4
2-1 人行道相關使用材料 4
2-2 人行道評估指標 4
2-2-1 舒適性指標 4
2-2-2 人因危害指標 5
2-2-3 安全性指標 7
2-2-4 暢行性指標 8
2-3 人行道檢測儀器 8
2-3-1　履帶式儀器 8
2-3-2 滑軌式儀器 9
2-3-3 輪椅與車輪式儀器 9
2-3-4　檢測載具綜合比較 12
2-4 波譜分析 14
2-4-1 快速傅立葉轉換(Fast Fourier Transform, FFT) 15
2-4-2 希爾伯特黃轉換(Hilbert Huang Transform, HHT) 19
2-4-3 各文獻特徵參數選取討論 26
2-5 GEP模型建立 26
2-6 歐式距離分群 27
三、研究方法 29
3-1 儀器配置 32
3-2 先進人行環境多功能檢測儀儀器配置 34
3-2-1單點雷射 34
3-2-2 距離感測器 35
3-2-3 訊號擷取器 36
3-2-4 主機與系統整合 36
3-2-5 陀螺儀 36
3-2-6 雷射 37
3-3 舒適性指標模組建立 37
3-3-1　距離感測器驗證 38
3-3-2 位移感測器驗證 39
3-3-3 異常值(Outlier)剔除 41
3-3-4 中值濾波 42
3-3-5 儀器重複性與準確度驗證 42
3-3-6　WPRI值計算程式驗證 43
3-4人因危害指標模組建立 43
3-5 安全性指標模組建立 43
3-5-1 現地抗滑、紋理選取材料 44
3-5-2 快速傅立葉轉換 44
3-5-3 希爾伯特黃轉換 45
3-5-4 GEP 預測模型構建 45
3-6暢行性指標模組 46
3-7 人行道現地調查 47
3-7-1 測量鋪面地點 47
3-7-2 現地檢測程序 48
3-8　單因子變異數分析 49
四、實驗室驗證成果分析 50
4-1 距離感測器驗證結果 50
4-2 位移感測器驗證結果 50
4-2-1 雷射電壓轉高度校正 50
4-2-2 單點雷射量測驗證 51
4-3　連續量測驗證 52
4-3-1 中值濾波 53
4-3-2 快速傅立葉轉換分離雜訊 55
4-3-3 儀器重複性與準確度驗證 56
4-3-4 濾波成效驗證 56
五、舒適性指標 57
5-1 剖面訊號處理及分析 57
5-1-1 剖面FFT分析 58
5-1-2 剖面HHT分析 60
5-2 現地WPRI測量 62
5-3 現地WPRI無母數單因子變異數分析 66
5-4 舒適性指標成效評估 67
六、人因危害指標 69
6-1 現地振動量測 69
6-2 現地振動無母數單因子變異數分析 71
6-3 人因危害指標成效評估 73
6-4 現地WPRI和振動資料綜合比較 73
七、安全性指標 75
7-1 現地抗滑量測 75
7-2 使用FFT波型分解進行GEP建模預測 76
7-3 使用HHT波型分解進行GEP建模預測 80
7-4 安全性指標成效評估 83
八、暢行性指標 85
8-1 人行道設施物建模 85
8-1-1 點雲資料校正 86
8-1-2 歐氏距離分群噪點濾除 86
8-1-3陀螺儀姿態、距離感測器資料紀錄 87
8-1-4 全斷面點雲擬合成果 88
8-2 路徑淨寬掃描 91
8-3暢行性指標成效評估 92
九、結論與建議 93
9-1　結論 93
9-1-1儀器建置成果 93
9-1-2舒適性、人因危害指標 93
9-1-3安全性指標 94
9-1-4 圖資建模與暢行性指標 94
9-1-5儀器建置成果綜合性評估 95
9-2 建議 95
參考文獻 97
附錄一、各市檢測街廓路段 102
附錄二、現地抗滑選取材料照片 108
附錄三、紋理FFT分析 119
附錄四、紋理HHT分析 137

參考文獻

[1] 衛生福利部統計處，2023，「2.3.3身心障礙者人數按等級及類別分」。
[2] 曾佳馨，2021，「以ASTM WPRI與垂直加速度建構檢測設備與評估人行環境舒適性之初步研究」，碩士論文，國立中央大學。
[3] 交通部，2023，「道路交通管理處罰條例」。
[4] 內政部國土管理署，2022，「市區道路及附屬工程設計規範」。
[5] ISO 2631-1, (2010). “Mechanical vibration and shock — Evaluation of human exposure to whole-body vibration — Part 1: General requirements. Standard – amendment 1, No. 2631-1/AMD 1:2010,” International Organization for Standardization.
[6] ASTM E3028-23. (2023). “Standard Practice for Computing Wheelchair Pathway Roughness Index as Related to Comfort, Passability, and Whole Body Vibrations from Longitudinal Profile Measurements. Standard”, American Society for Testing and Materials.
[7] ASTM E303-22. (2022). “Standard Test Method for Measuring Surface Frictional Properties Using the British Pendulum Tester”, American Society for Testing and Materials.
[8] 「CNS 16106人行面磚防滑性試驗法－濕式擺錘法」，2019，經濟部標準檢驗局。
[9] 交通部，2021，「交通工程規範」。
[11] KS F 2530:2018, (2018). “Standard specifications for stone block foe pavement”, Korean Standards Association (KSA).
[11] Jing, W., Shiying S., and Xiaoquang, Z., (2015). “SUnstructured road detection and path tracking for tracked mobile robot,” The 5th Annual IEEE International Conference on Cyber Technology in Automation, China.
[12] Timothy, M., Daniel, S., Laith, T., Nader, T., and Hussain, U.B., (2012) “Characterization of Asphalt Pavement Surface Texture,” Transportation Research Record: Journal of the Transportation Research Board, No. 2295, Transportation Research Board of the National Academies, Washington, D.C., 2012, pp. 19–26.
[13] Duvall, J. (2013). “Development of Surface Roughness Standard for Wheelchair Pathways,” University of Pittsburgh, U.S.A.
[14] ASTM E1927. (2018). “Standard Guide for Conducting Subjective Pavement Ride Quality Ratings.”, American Society for Testing and Materials.
[15] Sinagra, E.J. (2014). “The design, development, characterization, and validation of a pathway measurement tool,” Duquesne University, U.S.A.
[16] Duvall, J., Sinagra, E., Cooper, R., and Pearlman, J. (2016). “Proposed Pedestrian Pathway Roughness Thresholds to Ensure Safety and Comfort for Wheelchair Users,” The Official Journal of RESNA, RESNA, Vol. 28, No. 4, pp. 209–215.
[17] Timothy, M., Daniel, S., Laith, T., Nader, T., and Hussain, U.B., (2012) “Characterization of Asphalt Pavement Surface Texture,” Transportation Research Record: Journal of the Transportation Research Board, No. 2295, Transportation Research Board of the National Academies, Washington, D.C., 2012, pp. 19–26.
[18] Duvall, J. (2013). “Development of Surface Roughness Standard for Wheelchair Pathways,” University of Pittsburgh, U.S.A.
[19] 尹赫，(2017)，“Evaluation of Wheelchair Pathway Serviceability in Taiwan”，碩士論文，國立台灣大學。
[20] Sinagra, E.J. (2014). “The design, development, characterization, and validation of a pathway measurement tool,” Duquesne University, U.S.A.
[21] Michael W. Sayers, and Steven M. Karamihas, (1998) “The little book of profiling: basic information about measuring and interpreting road profiles”, University of Michigan
[22] Ueckermann, A., Wang, D., Oeser, M., & Steinauer, B. (2015). A contribution to non-contact skid resistance measurement. International Journal of Pavement Engineering, 16(7), 646-659.
[23] Michael W. Sayers, and Steven M. Karamihas, (1998) “The little book of profiling: basic information about measuring and interpreting road profiles”, University of Michigan
[24] Ueckermann, A., Wang, D., Oeser, M., & Steinauer, B. (2015). A contribution to non-contact skid resistance measurement. International Journal of Pavement Engineering, 16(7), 646-659.
[25] Zhan, Y., Liu, C., Deng, Q., Feng, Q., Qiu, Y., Zhang, A., & He, X. (2022). Integrated FFT and XGBoost framework to predict pavement skid resistance using automatic 3D texture measurement. Measurement, 188, 110638.
[26] Zhu, X., Yang, Y., Zhao, H., Jelagin, D., Chen, F., Gilabert, F. A., & Guarin, A. (2021). Effects of surface texture deterioration and wet surface conditions on asphalt runway skid resistance. Tribology International, 153, 106589.
[27] Serigos, Pedro A., Andre De Fortier Smit, and Jorge A. Prozzi., (2014), "Incorporating surface microtexture in the prediction of skid resistance of flexible pavements.", Transportation Research Record 2457.1: 105-113.
[28] Kane, M. (2022). Determining the Prominence of Texture Scales on Road Skid Resistance. In Proceedings of the RILEM International Symposium on Bituminous Materials: ISBM Lyon 2020 1 (pp. 553-558). Springer International Publishing.
[29] Can Chen, Haoyuan Luo, Siyu Chen, Yangzezhi Zheng, Xiyin Liu, Tao Ma, Leyi Zhu. (2023). “Exploring the relationship between pavement surface texture and friction based on Hilbert-Huang Transformation.“, Tribology International, 187, 108737.
[30] Yuan-shuai Dong, Yun Hou, Jia-lei Tian, Yu-xuan Cao, Chen-wei Guo, Tuo Fang, and Jing Zhou. (2021). “Hilbert-Huang Transform in Pavement Texture and Skid-Resistance Study.”, In International Conference on Green Intelligent Transportation System and Safety (pp. 341-353). Singapore: Springer Nature Singapore.
[31] WENYING YU., (2019), “Characterizing Pavement Skid Resistance for Roadway Crash Prediction in Oklahoma”, Doctoral dissertation, Oklahoma State University.
[32] Miao Yu1, Yao Kong, Chuanhai Wu, Xinquan Xu, Shanqiang Li, Haifeng Chen, Lingyun Kong. (2021), “The effect of pavement texture on the performance of skid resistance of asphalt pavement based on the Hilbert-Huang transform.”, Arabian Journal for Science and Engineering, 46, 11459-11470.
[33] Nicolas Gagarin, Norden E. Huang, Morton S. Oskard, Dennis G. Sixbey, James R. Mekemson, (2004), “The application of the Hilbert-Huang transform to the analysis of inertial profiles of pavements.”, International journal of vehicle design, 36(2-3), 287-301.
[34] Candida Ferreira, “Gene Expression Programming Mathematical Modeling by an Artificial Intelligence 2nd ed. Springer, 2006.
[35] Jun Liu, Kezhen Yan, Lingyun You, Pei Liu & Kezhen Yan., (2017), "Prediction models of mixtures′ dynamic modulus using gene expression programming." International Journal of Pavement Engineering, 18(11), 971–980.
[36] Mehran Mazari, Daniel D. Rodriguez, (2016), "Prediction of pavement roughness using a hybrid gene expression programming-neural network technique", Journal of Traffic and Transportation Engineering,3 (5): 448–455.
[37] Ashwini R. Tenpe and Anjan Patel, (2018), "Application of genetic expression programming and artificial neural network for prediction of CBR". Road Materials and Pavement Design, v.19(1), pp.1–18,.
[38] Kaffayatullah Khan, Babatunde Abiodun Salami, Arshad Jamal, Muhammad Nasir Amin, Muhammad Usman, Majdi Adel Al-Faiad, Abdullah M. Abu-Arab and Mudassir Iqbal, (2022), “Prediction Models for Estimating Compressive Strength of Concrete Made of Manufactured Sand Using Gene Expression Programming Model.”, Materials, 15.17: 5823.
[39] Rana Imam, Yasmin Murad, Ibrahim Asi, Anis Shatnawi, (2021)ㄝ"Predicting pavement condition index from international roughness index using gene expression programming. "Innovative Infrastructure Solutions, 6: 1-12.
[40] Linyi Yao, Zhen Leng, Jiwang Jiang, Fujian Ni, Zili Zhao, (2021) "Nondestructive prediction of rutting resistance of in-service middle asphalt layer based on gene expression programming." Construction and Building Materials, 293: 123481.
[41] C.D. Lewis, Industrial and business forecasting methods, Butterworth- Scientific, 1982.
[42] Shih-Huang Chen, Cheng-Kai Huang, Dita Adelafani, Yi-Yang Cheng, (2023). The preliminary study of the forecasting model between surface texture and various material parameters for the ISO 10844 test track. Case Studies in Construction Materials, 19, e02523.
[43] D. Kong and W. Xy, ‘‘Vehicle target identification algorithm based on poin cloud of vehicle 32-line laser lidar,’’ Sci. Technol. Eng., vol. 18, no. 5, pp. 81–85, May 2018.
[44] X. Z. Wang, J. Li, H. J. Li, and B. X. Shang, ‘‘Obstacle detection based on 3D laser scanner and range image for intelligent vehicle,’’ J. Jilin Univ. (Eng. Technol. Ed.), vol. 46, no. 2, pp. 360–365, Apr. 2016.
[45] S. I. R. Rodríguez and F. de Assis Tenorio de Carvalho, ‘‘Fuzzy clustering algorithm based on adaptive Euclidean distance and entropy regularization for interval-valued data,’’ presented at the Artif. Neural Netw. Mach. Learn., Oct. 2018.
[46] L. Bai, L. Yan, and Z. M. Ma, ‘‘Determining topological relationship of fuzzy spatiotemporal data integrated with XML twig pattern,’’ Appl. Intell., vol. 39, no. 1, pp. 75–100, 2013.
[47] D. H. Zhu, ‘‘Euclidean cluster extraction in PCL,’’ in Point Cloud Library PCL Study Tutorial, 2th ed. Beijing, China: Univ. Aeronautics and Astronautics Press, 2012, pp. 338–350.
[48] H. Rashmanlou, S. Sahoo, R. A. Borzooei, M. Pal, and A. Lakdashti, ‘‘Computation of shortest path in a vague network by Euclidean distance,’’ J. Multiple-Valued Logic Soft Comput., vol. 30, no. 1, pp. 115–123, 2018.
[49] ISO 13473-3, (2002). “Characterization of pavement texture by use of surface profiles — Part 3: Specification and classification of profilometers”, International Organization for Standardization.
[50] 「CNS 15046慣性剖面儀量測舖面縱向剖面試驗法」，2006，經濟部標準檢驗局。
[51] 「CNS 15371鋪面量測用慣性剖面儀驗證法」，2010，經濟部標準檢驗局。
[52] ISO 13473-1, (2019), “Characterization of pavement texture by use of surface profiles Part 1: Determination of mean profile depth”, International Organization for Standardization.
[53] ASTM E178-21, (2021) “Standard Practice for Dealing With Outlying Observations”, American Society for Testing and Materials.
[54] 內政部國土管理署，「112年度市區道路養護管理暨人行環境無障礙考評計畫評鑑報告(附錄E街廓考評範圍說明)」，社團法人鋪面工程學會。
[55] 郭欽培. (1995). 無障礙設計之建築觀: 從殘障者的特性與需求談起. 詹氏。
[56] 黃振愷，2018，「臺灣六都市區道路平坦度分析與改善策略建議之研究」，碩士論文，國立中央大學
[57] ISO 2631-5:2018, (2018), “Mechanical vibration and shock Evaluation of human exposure to whole-body vibration Part 5: Method for evaluation of vibration containing multiple shocks”, International Organization for Standardization.

指導教授

陳世晃(Shih-Huang Chen)

審核日期

2024-1-29

推文