臺灣博碩士論文加值系統

本論文為自動駕駛高爾夫球車之決策分析與系統驗證，首先於高爾夫球車平台上建構自駕系統，對電門、轉向與剎車系統進行電控改裝，並建立一車資通訊系統作為控制訊號接收端，並透過CAN bus傳送對應訊號至各控制電路，配合感知系統，如光學雷達、慣性偵測單元與視覺鏡頭等感知器辨識周遭環境特徵，並透過決策系統進行路徑上的規劃，最終傳至車輛上的車資通訊系統進行控制，並透過微控制器控制轉向馬達與高爾夫球車驅動馬達，使高爾夫球車完成場域內自動駕駛。
首先使用機器人作業系統(Robot Operating System, ROS)作為上端決策系統，透過光學雷達進行偵測，透過SLAM(Simultaneous Localization And Mapping)完成場域內的地圖建置，搭配慣性偵測單元能夠精準定位車輛位置，之後使用Waypoint Follower演算法繪製出行車路徑，並在車輛行駛過程中，利用鏡頭即時擷取高爾夫球車前方路況，並對圖像資料進行邊緣檢測(Edge Detection)與霍夫轉換(Hough Transform)等演算處理，找出車道線位置，使高爾夫球車於Waypoint Follower的行車路徑中進行車道修正，行駛於可行駛範圍內，讓高爾夫球車能夠完成測試情境場域無人駕駛，透過系統驗證，進行實際場域內的測試與評估。

This thesis centers around the decision making and system verification of a self-driving golf cart. Initially, a self-driving system is established on the golf cart platform, with electronic modifications implemented on the accelerator, steering, and braking system. A Vehicle Control Unit is created as the receiver of control signals. The corresponding signals are transmitted to each control circuit via the CAN bus system, with a perception system that includes devices such as optical radar, inertial detectors, and vision cameras, used to identify the characteristics of the surrounding environment. Through the decision-making system, route planning is carried out, and eventually transmitted to the VCU on the vehicle for control. Steering motors and body motors are controlled via a microcontroller, enabling the golf cart to complete autonomous driving within the field.
The Robot Operating System (ROS) is utilized as the upper-level decision-making system. With optical radar for detection, a map within the field is built using SLAM (Simultaneous Localization And Mapping), combined with an Inertial Measurement Unit to precisely locate the vehicle's position. The Waypoint Follower algorithm is subsequently used to plot driving routes. In the course of the vehicle's journey, a camera captures real-time conditions in front of the golf cart, applying edge detection and Hough Transform to image data to locate lane lines, enabling the golf cart to make lane adjustments within the driving route designed by the Waypoint Follower. This ensures the golf cart stays within drivable boundaries, facilitating unmanned driving within designated areas. Through system verification, actual field tests and evaluations are conducted, comparing the autonomous driving performance under different control strategies.

摘要......i
Abstract......ii
誌謝......iii
目錄......iv
表目錄......vi
圖目錄......vii
第一章 緒論......1
1.1 前言......1
1.2 研究動機......2
1.3 研究目的......3
1.4 研究方法......3
1.5 文獻回顧......4
1.5.1自駕車發展......4
1.5.2自駕車感知決策系統......5
1.5.3 SLAM演算法......6
1.5.4 車道線檢測演算法......7
1.6 論文架構......7
第二章 高爾夫球車自駕系統建立......8
2.1 車身系統架構......8
2.2 車資通訊系統......9
2.3 電控轉向系統......12
2.4 電門控制系統......15
2.5 電控剎車系統......18
2.6 機器人作業平台建立......20
第三章 高爾夫球車感知決策系統建立......24
3.1 機器視覺感知系統......25
3.1.1影像前處理......26
3.1.2特徵擷取......27
3.1.3邊緣檢測......29
3.1.4霍夫變換法......31
3.2 光學雷達系統......32
3.3 自駕決策之設計......35
3.3.1車道維持系統......36
3.3.2 Waypoint Follower路徑跟隨......38
第四章 實驗方式與結果......41
4.1自駕系統驗證......41
4.2 Lane Keeping驗證......46
4.2.1偏車道起步......46
4.2.2彎道測試......47
4.3實車決策系統驗證......49
第五章 結論與未來工作......51
5.1結論......51
5.2未來工作......51
參考文獻......53
附錄......55
A.車資通訊系統......55
設定區......55
控制訊號發送......56
回授訊號擷取......56
B.電控轉向系統......57
設定區......57
中斷觸發......58
控制訊號擷取......58
轉向控制......58
C.電門控制系統......59
設定區......59
控制訊號擷取與電門配值......60
檔位切換......60
車速回授計算與傳送......61
電門控制(DAC Control)......61
D.電控剎車系統......62
設定區......62
作動控制......62
控制訊號擷取與斷訊邏輯......63
Extended Abstract......64

[1]施泓銘，2022，“自駕車法規趨勢與實車驗證評價”，ARTC車研專刊，頁56~65。
[2]李易俊，2022，“自動駕駛運用發展與自駕整合決策技術”，2022車輛研測專刊,，頁38~43。
[3]陳敬典，2020，“全球自駕車產業發展現況與未來趨勢”，2020車輛研測專刊，頁19~28。
[4]許立佑，2020，“車輛環境感測技術”，2020車輛研測專刊，頁31~42。
[5]I. Filip, J. Pyo, M. Lee and H. Joe, "Lidar SLAM Comparison in a Featureless Tunnel Environment," 2022 22nd International Conference on Control, Automation and Systems (ICCAS), Jeju, Korea, Republic of, 2022, pp. 1648-1653, doi: 10.23919/ICCAS55662.2022.10003820.
[6]T. Shan, B. Englot, D. Meyers, W. Wang, C. Ratti and D. Rus, "LIO-SAM: Tightly-coupled Lidar Inertial Odometry via Smoothing and Mapping," 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Las Vegas, NV, USA, 2020, pp. 5135-5142, doi: 10.1109/IROS45743.2020.9341176.
[7]Aminuddin, N. S., Mat Ibrahim, M., Mohd Ali, N., Ahmad Radzi, S., Mohd Saad, W. H., & Darsono, A. M. (2017). A NEW APPROACH TO HIGHWAY LANE DETECTION BY USING HOUGH TRANSFORM TECHNIQUE. Journal of Information and Communication Technology, 16(2), 244–260. Retrieved from
[8]王崑勝，2022，“高爾夫球車線控系統整合設計與分析”，7月。
[9]Arduino, “Arduino Hardware,” ,https://www.arduino.cc/en/hardware.
[10]autowp, “arduino-mcp2515,” 2016. ,https://github.com/autowp/arduino-mcp2515.
[11]RobTillaart, “MCP4725,” ,https://github.com/RobTillaart/MCP4725.
[12]張元泰&莊凱翔，2015，“具低成本, 高速電動車馬達系統之電流式磁感測器設計”，工業技術研究院電腦與通訊工業研究所163期，頁65~70。
[13]Nvidia, “Nvidia.com,” Nvidia, https://www.nvidia.com/zh-tw/autonomous-machines/embedded-systems/jetson-agx-xavier/.
[14]ROS，“機器人作業平台”，https://www.ros.org/.
[15]王立安，2022，“基於模糊控制時線車道變換之自動駕駛”，頁18。
[16]raja_961, “Autonomous Lane-Keeping Car Using Raspberry Pi and OpenCV,” https://www.instructables.com/Autonomous-Lane-Keeping-Car-Using-Raspberry-Pi-and/.
[17]opencv,“Open Source Computer Vision Library,” 2020. https://github.com/opencv/opencv.
[18]S. Z,“Vision Based Lane Detection for Self-Driving Car,”IEEE,2020.
[19]v.-a. GrantMeStrength, “RGB 色彩空間,” microsoft, 13 06 2023. https://learn.microsoft.com/zh-tw/windows/win32/wcs/rgb-color-spaces.
[20]M. Sharifi; M. Fathy; M.T. Mahmoudi, “A classified and comparative study of edge detection algorithms,” 於 IEEE, Las Vegas, NV, USA, 2002.
[21]Dr.S.Vijayarani, Mrs.M.Vinupriya, “Performance Analysis of Canny and Sobel Edge Detection Algorithms in Image Mining,” 2013.
[22]蔡承運，2019，“具即時車道線辨識之三角履帶自走車開發”。
[23]林禹任，2020，“電動車主動轉向於車道維持禹追跡控制之研究”。
[24]TixiaoShan, LIO-SAM,2020.https://github.com/TixiaoShan/LIO-SAM.
[25]黃世杰，“自駕車實車功能驗證測試方法”。