居家服務型機器人人臉追蹤

A Home Service Robot with Face Tracking

詹勳宇(CHAN,HSUN-YU)

指導教授 ： 蕭俊祥

國立臺北科技大學/機電學院/機電整合研究所/碩士(2011年)

摘要

本論文以德國FESTO公司所開發的Robotino全向輪機器人為移動平台，並在Visual C++上搭配OpenCV函式實現人臉偵測，並取得人臉中心x座標值，將座標資訊傳送至LabVIEW，再根據系統的誤差，利用比例、微分控制運算出控制量，進而決定機器人左右旋轉角度。同時搭配雷射測距儀，掃描人體距離並當作模糊邏輯的輸入變數，經由模糊邏輯的推論可以得到機器人的速度，方可使機器人達到定距離的追蹤。再以紅外線感測器進行必要之閃避障礙物。避障模式是利用全向輪特性，一邊將鏡頭對準人臉，一邊使機器人與障礙物保持等距，對準人臉的目的在於確保隨時都能即時人臉追蹤。

關鍵字

人臉偵測 ； PD控制器 ； 模糊邏輯控制

並列摘要

In this study, Robotino, which is an omni-directional wheeled robot and developed by German FESTO company, is used to be the moving platform. Face detection algorithm is achieved by coding Visual C++ program and using the OpenCV functions. When the x coordinate value of the face center is obtained, the coordinate information is fedback to controller implemented in LabVIEW software. According to the deviation error of the face from the image center, a PD controller is used to calculate the control signal for suitably rotating the Robotino robot. In the meantime, a laser range finder is employed to detect the distance of human body which will be used as the input variable of fuzzy logic controller. The fuzzy logic controller can determine the robot speed so that cruise control can be obtained. In addition, obstacle avoidance is carried out by using the robot infrared sensors. Obstacle Avoidance algorithm is so designed that the robot always tracks the face and simultaneously wall follows the obstacle.

並列關鍵字

Face Detection ； PD Controller ； Fuzzy Logic Control

參考文獻

[4]Yang, M. H. Kriegman and D. Ahuja,”N.: Detecting faces in images: A survey”. IEEE Transactions on Pattern Analysis and Machine Intelligence 24(1) (2002) 34–58H.

[7]T. Yata, L. Kleeman, and S. Yuta, "Wall following using angle information measured by a single ultrasonic transducer," in IEEE International Conference on Robotics and Automation, 1998, pp. 1590-1596.

[8]P. Viola and Michael J. Jones. Rapid Object Detection using a Boosted Cascade of Simple Features. IEEE CVPR, 2001.

[14]L. A. Zadeh, “Soft computing and fuzzy logic,” IEEE Softw, vol. 11,no. 6, pp. 48–56, Nov. 1994.

[15]P. Taguchi and M. Araki, “On tuning of two-degree-of-freedom PID controllers with consideration on location of disturbance input,” Trans. SICE, vol. 38, pp. 441-446, 2002.