本論文旨在應用Kinect實現物體搬運之移動機器人，在室內環境中對特定目標物的辨識與搬運至指定位置。建立已知的室內地圖，計算移動機器人與目標物間的軌跡，並以室內地圖檢測軌跡的安全性，使移動機器人到達適當進行辨識目標物的地點。然後，利用Kinect讀取場景影像，對深度影像的破碎進行修補，接著使用相機參數轉換二維影像至點雲資訊，藉由前處理刪除點雲與物件分群，再以目標物的色彩資訊與幾何特徵進行辨識，計算移動機器人與目標物於相機座標中相對角度與距離。最後，移動機器人以FPGA發展控制模組，達到步進馬達位置及速度控制與伺服馬達的角度控制，使移動機器人能夠進行移動及夾爪抬升及閉合等動作，實現對目標物的夾取與搬運。

This thesis aims to implement an object-handling robot system, which is equipped with a Kinect sensor, so that the mobile robot can identify ae specific target and moves to a designated location in the indoor environment. First, one establishes a known indoor map and input the target location. Then the robot can use the indoor map to detect the collision-free trajectory, so that the mobile robot reaches the properly identified position. Kinect sensor is used to capture the scene image. First, repairing the depth image which has the unknown depth value. Next, converting the 2D image to the point clouds information by camera parameters. And then, deleting point clouds and object grouping by pre-processing. Finally, the color information and geometric features used to identify the target and calculate the relative angle and distance between the target and the mobile robot in the Kinect sensor coordinates. Mobile robot develops control module with FPGA in order to achieve position control and speed control of stepper motor and angle control of servo motor. The mobile robot realizes the grabbing and handling of the target by movement and jaw lifting and closing.

[1] Pramod Ranade, "Linear motor control without the math," SPJ Embedded
Technologies, APRIL 01, 2009.
[2] Krishna Rao Vijayanagar, Maziar Loghman and Joohee Kim, " Real-time
refinement of Kinect depth maps using multi-resolution anisotropic diffusion," Springer Science + Business Media, pp. 414-425, New York 2013.
[3] Na-Eun Yang, Yong-Gon Kim and Rae-Hong Park, "Depth hole filling using the depth distribution of neighboring regions of depth holes in the Kinect Sensor," IEEE International Conference on Signal Processing, Communication and Computing (ICSPCC 2012), pp. 658-661, 2012.
[4] Daniel Herrera C., Juho Kannala, and Janne Heikkil¨a," Joint depth and color
camera calibration with distortion correction ", IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 2058 – 2064, 2012.
[5] Mauro Bellonea, Luca Pascalib and Giulio Reina, "A Kinect-based parking
assistance system," Department of Engineering for Innovation, Università del Salento, Via Arnesano, 73100 Lecce, pp. 127-140, Italy, 2013.
[6] Hui Tan, Mingliang Huang, Qi Zhang; Yi Liu, "Controlling system design of high
speed stepping motor based on serial communication," 4th IEEE International
Symposium on Microwave, Antenna, Propagation and EMC Technologies for
Wireless Communications, pp. 282 – 285, 2011.
[7] D. Carrica, M. A. Funes, S. A. Gonzalez, "Novel stepper motor controller based on
FPGA hardware implementation," IEEE/ASME Transactions on Mechatronics,
vol. 8, no. 1, pp. 120 – 124, 2003.
[8] Jose I. Quinones, "Applying acceleration and deceleration profiles to bipolar
stepper motors," TI Applications Engineer, 2012.

[9] Priyanka Sudhakara,Velappa Ganapathy and Karthika Sundaran, "Probabilistic
roadmaps-spline based trajectory planning for wheeled mobile robot,"
International Conference on Energy, Communication, Data Analytics and Soft
Computing (ICECDS), pp. 3579 – 3583, 2017.
[10] Priyanka Sudhakara, Velappa Ganapathy and Karthika Sundaran, "Mobile robot
trajectory planning using enhanced artificial bee colony optimization algorithm,"
IEEE International Conference on Power, Control, Signals and Instrumentation
Engineering (ICPCSI), pp. 363 – 367, 2017.
[11] Jaesik Park, Hyeongwoo Kim, Yu-Wing Tai, Michael S. Brown and In So
Kweon, "High-quality depth map upsampling and completion for RGB-D
Cameras," IEEE Transactions on Image Processing, pp. 5559 – 5572, 2014.
[12] Chongyu Chen, Jianfei Cai, Jianmin Zheng, Tat-Jen Cham and Guangming Shi,
"A color-guided, region-adaptive and depth-selective unified framework for
Kinect depth recovery," IEEE 15th International Workshop on Multimedia Signal
Processing (MMSP), pp. 007 – 012, 2013.
[13] Jan Smisek, Michal Jancosek and Tomas Pajdla, "3D with Kinect", IEEE
International Conference on Computer Vision Workshops (ICCV Workshops),
pp. 1154 – 1160, 2011.
[14] Z. Zhang, "A flexible new technique for camera calibration," IEEE Transactions
on Pattern Analysis and Machine Intelligence, vol. 22, no. 11, pp. 1330 – 1334,
2000.
[15] Oliver Wasenmüller, Marcel Meyer and Didier Stricker, "CoRBS: comprehensive RGB-D benchmark for SLAM using Kinect v2", IEEE Winter Conference on Applications of Computer Vision (WACV), 2016.
[16] Aitor Aldoma, Zoltan-Csaba Marton, Federico Tombari, Walter Wohlkinger, Christian Potthast, Bernhard Zeisl, Radu Bogdan Rusu, Suat Gedikli and Markus Vincze, "Tutorial: point cloud library: three-dimensional object recognition and 6 DOF pose estimation," IEEE Robotics & Automation Magazine, pp. 80 – 91, 2012.
[17] Radu Bogdan Rusu, Steve Cousins, "3D is here: point cloud library (PCL) ,
" IEEE International Conference on Robotics and Automation, pp. 1-4, 2011.
[18] Giulio Reina, Rocco Galati and Annalisa Milella, "All-terrain estimation for mobile robots in precision agriculture," IEEE International Conference on Industrial Technology (ICIT), pp. 63-68, 2018.
[19] Hongpeng Yin, Yi Chai, Simon X. Yang and Gauri S. Mittal, "Ripe tomato recognition and localization for a tomato harvesting robotic system," International Conference of Soft Computing and Pattern Recognition, pp. 557 – 562, 2009.
[20] 林佳蓁，以二維影像特徵作為ICP初始估測之三維點雲對位，碩士論文，國立雲林科技大學，2015。
[21] 賴柏宏，基於三維影像之坦克車即時姿態估測研究，碩士論文，國立交通大學，2012。
[22] 施慶隆、李文猶，機電整合與控制--多軸運動設計與應用，第三版，全華書局股份有限公司，2015。