由於衛星資料與地圖詳細資訊的取得不易，使得一般人無法輕易使用現實世界中的行車路線，作為自駕車控制相關領域的驗證。因此，本論文提出了一種能提供給自駕車系統/模擬器控制的軌跡產生器，能大幅降低實際路線驗證的難易度。所提技術包含一、以Google Maps的路線導航圖為圖資，開發了一種基於影像處理的路徑轉換器，使得具有比例約為2.20公尺／像素的路線資訊影像能夠轉換成具有以精確浮點數表示之位置的軌跡。二、結合傳統路徑規劃演算法與均值平滑化，提出一種能使路徑更加平滑之參數化平滑法。三、為使轉換後之路徑能應用於自駕車系統/模擬器之控制，規劃出一種符合於該線段曲率的離線多參數速度設計器。相對應於所提技術一、在雙向道路中，提供使整段或部分路徑向右平移之功能。二、比較所提出之方法與僅使用傳統路徑規劃演算法或均值平滑化的差異。三、透過最高速限、轉彎速限與加速度等可人為調整之參數，產生出一組平滑的速度數據，並於上述路徑上進行採樣，得到一條可供控制系統直接使用的軌跡。此外，透過基於Vehicle Dynamics-MATLAB & Simulink-MathWorks 2018a的控制研究做為驗證工具，證實此方法所產生的軌跡，作為控制系統中被追蹤之參考訊號是可行的。

Due to the difficulty to obtain the detailed information of satellite maps, a navigation route for autonomous vehicles to driving in the real world cannot be easily obtained to verify the control theory in the field of self-driving vehicles. Therefore, this thesis proposes a trajectory generator that can be provided for the control of the autonomous vehicle and simulator. It can greatly reduce the difficulty to verify the control theory for autonomous vehicles driving on actual route. This includes: (i) A path converter to convert the marked route on Google Maps image in a scale of 2.20 meters per pixel into a trajectory in position represented by the accurate floating number, (ii) A parametric smoothing method to provide a smooth path by combining traditional path planning algorithms and the average smoothing method, and (iii) An off-line multi-parameter velocity generator to make the converted path applicable to the control of the self-driving vehicle and simulator. Consequently, (i) If there is only a single lane in a single direction on the road, it also has the function to shift the path to the right; (ii) The difference between traditional path planning algorithms, average smoothing methods, or the mixture method is also shown in this thesis; (iii) Based on the curvature, maximum speed limit, turning speed limit, and acceleration, engineer can design a set of smooth velocity data. The trajectory that is re-sampled on the above-mentioned path with the set of smooth velocity data can be used directly for the control system. In addition, the path-tracking control for the customized vehicle supported by Vehicle Dynamics-MATLAB & Simulink-MathWorks 2018a is used as a verification to verify that the trajectory generated by this method is feasible as a reference signal to be tracked in the control system.

[1] Amidi, O., “Integrated mobile robot control”, Technical Report CMU-RI-TR-90-17, Robotics Institute, Carnegie Mellon University, Pittsburgh, Pennsylvania, May 1990.
[2] Atherton, T.J. and Kerbyson, D.J., “Size invariant circle detection,” Image and Vision Computing, vol. 17, pp. 795-803, Sept. 1999.
[3] Bradski, G. and Kaehler, A., Learning OpenCV: Computer Vision with the OpenCV Library, O'Reilly Media, Sebastopol, California, United States, 2008.
[4] Bresenham, J.E., “Algorithm for computer control of a digital plotter”, IBM Systems Journal, vol. 4, pp. 25-30, 1965.
[5] Canny, J., “A computational approach to edge detection”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-8, pp. 679-698, 1986.
[6] Chu, W.J., Prototype and Verification of Path-Tracking-Oriented Simulators for Customized Autonomous Vehicle, Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan, R.O.C., Thesis for Master, July 2020.
[7] Coulter, R. C., “Implementation of the pure pursuit path tracking algorithm,” Technical Report CMU-RI-TR-92-01, Robotics Institute, Carnegie Mellon University, Pittsburgh, Pennsylvania, Jan. 1992.
[8] Dubins, L.E., “On curves of minimal length with a constraint on average curvature, and with prescribed initial and terminal positions and tangents,” American Journal of Mathematics, vol. 57, no. 3, pp.497-516, July 1957.
[9] Ferguson, D. and Stentz, A., “Using interpolation to improve path planning: The field D* algorithm,” Journal of Field Robotics, vol. 23, pp. 79-101, 2006.
[10] Haralick, R.M. and Shapiro, L.G., Computer and Robot Vision, vol. 1, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, United States, 1992.
[11] Hyndman, R.J., “Moving averages,” International encyclopedia of statistical science, Springer, Berlin, Heidelberg, pp. 866-869, 2011.
[12] Kao, C., Path-Tracking-Oriented Simulators for Customized Autonomous Vehicles, Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan, R.O.C., Thesis for Master, July 2019.
[13] Kong, T.Y. and Rosenfeld Azriel, Topological Algorithms for Digital Image Processing, Elsevier Science, Inc., Amsterdam, Netherlands, 1996.
[14] Krogh, B. and Thorpe, C., “Integrated path planning and dynamic steering control for autonomous vehicles,” 1986 IEEE International Conference on Robotics and Automation, 1986.
[15] Lam, L., Lee, S.W., and Suen, C.Y., “Thinning methodologies-A comprehensive survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol 14, no. 9, pp. 869-885, September 1992.
[16] Liu, B.L., Prototype and Verification of Advanced Control Technique for Autonomous Vehicle, Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan, R.O.C., Master Thesis, July 2020.
[17] Pratt, William K., Digital Image Processing: PIKS SCIENTIFIC Inside, John Wiley & Sons, Inc., Hoboken, New Jersey, United States, 2007.
[18] Reeds, J. and Shepp, L., “Optimal paths for a car that goes both forwards and backwards,” Pacific Journal of Mathematics, vol. 145, no. 2, pp. 367-393, 1990.
[19] Shin, D.H., Singh, S., and Lee, J.J., “Explicit path tracking by autonomous vehicles,” Robotica, vol. 10, pp. 539-554, 1992.
[20] Spong, M.W., Hutchinson, S., and Vidyasagar, M., Robot Modeling and Control, John Wiley & Sons, Inc., Hoboken, New Jersey, United States, 2006.
[21] Wang, C.H., Advanced Control Technique for Autonomous Vehicles, Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan, R.O.C., Master Thesis, July 2019.