本論文開發一套以三維影像資訊結合工業機械手臂之自動作業目標物選定與即時避障系統，藉由三維影像系統偵測物件資訊選定目標物件，並以物件表面法向量計算手臂運動中繼點與姿態，使機械手臂可成功完成作業任務。而當影像系統偵測出障礙物時，則藉由機械手臂與障礙物相對位置，決定機械手臂的閃避方向與閃避位移量，並基於順逆向運動學與機構特性簡化計算量提供即時閃避作動路徑，有效維持10Hz之即時運算作動能力。
本研究以上銀六軸機械手臂完成實機驗證三維影像物件偵測與機械臂即時避障作動路徑實現與控制性能。論文提出之方法可使機械臂成功閃避2m/s速度之動態障礙物，並達成疊合多物件取放任務。研究成果有助於智慧化視覺互動機械臂系統之實現與人機協同之彈性生產製造環境建構。

A Real-time obstacle avoidance robotic system using 3D visual guidance is proposed in this study. Targets information are detected by 3D image system for object recognition and positioning. A task motion relay point in the object surface normal direction is introduced to achieve the mission of pick and place. When an obstacle is detected, the relative position and distance of the arm to the obstacle is firstly determined by the 3D vision system. Subsequently, the obstacle avoiding motion planning is carried out according to kinematic relation intuitively. The updating performance of the proposed system is 10Hz.
An integrated 3D objects detection and real-time obstacle avoidance motion planning system proposed is implemented in associated with HIWIN RA605 six-axis robot to demonstrate and verify the real time performance of the system.

Beyeler, A., Zufferey, J. C., and Floreano, D. (2009). Vision-based control of near-obstacle flight. Autonomous robots, 27(3), 201-219.
Flacco, F., Kroeger, T., De Luca, A., & Khatib, O. (2015). A depth space approach for evaluating distance to objects. Journal of Intelligent & Robotic Systems, 80, 7.
Ghidary, S. S., Nakata, Y., Takamori, T., & Hattori, M. (2000). Human detection and localization at indoor environment by home robot. In Systems, Man, and Cybernetics, 2000 IEEE International Conference on. Vol. 2, 1360-1365.
Hernandez-Lopez, J. J., Quintanilla-Olvera, A. L., López-Ramírez, J. L., Rangel-Butanda, F. J., Ibarra-Manzano, M. A., & Almanza-Ojeda, D. L. (2012). Detecting objects using color and depth segmentation with Kinect sensor.Procedia Technology, 3, 196-204.
Munaro, M., Basso, F., and Menegatti, E. (2012). Tracking people within groups with RGB-D data. 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vilamoura.
Perdereau, V., Passi, C., & Drouin, M. (2002). Real-time control of redundant robotic manipulators for mobile obstacle avoidance. Robotics and Autonomous Systems, 41(1), 41-59.
Rakprayoon, P., Ruchanurucks, M., & Coundoul, A. (2011, December). Kinect-based obstacle detection for manipulator. In System Integration (SII), 2011 IEEE/SICE International Symposium on (pp. 68-73). IEEE.
Shen, H., Wu, H., Chen, B., Jiang, Y., & Yan, C. (2015). Obstacle avoidance algorithm for 7-DOF redundant anthropomorphic arm. Journal of Control Science and Engineering, 2015, 7.
Silberman, N., Hoiem, D., Kohli, P., and Fergus, R.. Indoor segmentation and support inference from RGBD images. (2012). Computer Vision -ECCV (European Conference on Computer Vision) 2012, 746-760, Florence, Italy.
Teichman, A., Lussier, J. T., and Thrun, S. (2013). Learning to Segment and Track in RGBD. 2013 IEEE Transactions on Automation Science and Engineering. 10(4), 841-852.
Zhang, H., Jia, Y., Xi, N., & Song, A. (2012, December). Obstacle avoidance for mobile manipulation by real-time sensor-based redundancy resolution. In Robotics and Biomimetics (ROBIO), 2012 IEEE International Conference on(pp. 2369-2374). IEEE.
陳冠宇、吳宗穎、俞牧之、曾乾特. (2011). 立體視覺伺服自主行動機器人之動態避障與控制. Journal of Advanced Engineering, 6(3), 161-167.
顧正偉. (2012).國家高速網路與計算中心-科學視算與互動媒體實驗室, http://viml.nchc.org.tw/.