基於Gazebo之人形機器人的崎嶇地形路徑規劃

Gazebo-based Path Planning of Uneven Terrain for Humanoid Robot

陳宗勳(Tsung-Hsun Chen)

指導教授 ： 李祖添 ； 劉智誠

淡江大學/工學院/電機工程學系碩士班/碩士(2019年)

https://doi.org/10.6846/TKU.2019.00466

摘要

本論文針對小型人形機器人，提出一個基於機器人作業系統(Robot Operating System, ROS)上，對於不平坦之複雜地形的路徑規劃方法，並實現於Gazebo模擬器之上。在影像處理方面，使用Gazebo提供之攝影機模組作為影像輸入，在經過色彩校正、過濾雜訊與物件分割得出感興趣之物件後，作為虛擬地圖資訊。在定位方面，使用自製機器人3D模組的位置作為地圖原點，判斷機器人在整個不平坦地形上的位置所在，同時以機器人的視覺影像來校正與不平坦地形的角度誤差。在導航方面，本論文提出一種對不平坦的複雜地形做出困難行走區域的判斷方法，使用A*演算法做路徑規劃，使機器人盡可能地避開困難區域。在實驗結果中可見，本論文所提出之方法能夠在已知環境中，避免機器人走在容易跌倒的高風險區域且順利抵達目的地。

關鍵字

中樞模式產生器 ； 複雜地形 ； A*演算法 ； Gazebo ； 人形機器人 ； 機器人作業系統

並列摘要

In this thesis, a path planning method for uneven terrain is proposed to be implemented on the Robot Operating System (ROS) and Gazebo simulator for a samll-size humanoid robot. In image processing system part, the Gazebo camera module is used as the image source. Color correction, filtering noise, and object segmentation method are used to mark interested objects as visual map information. In localization system part, the homemade robot 3D module is the map origin. Then these distances between the robot and features are determined. And the angular error of the uneven terrain with the vision of the robot are corrected. In navigation system part, this paper proposes a method to determine difficult walking area in the uneven terrain. Then A* algorithm is used to plan a path to make the robot avoid difficult walking area as much as possible. From the experimental results, the proposed method can avoid the robot walking into a high-risk area that is easy to fall, and arrive at the destination smoothly.

並列關鍵字

A* algorithm ； Central Pattern Generator ； Gazebo ； Humanoid Robot ； Robot Operating System (ROS) ； Uneven terrain

參考文獻

[1] “Robot.wiki.” URL: https://en.wikipedia.org/wiki/Robot.

[2] A.D. Kuo, “The relative roles of feedforward and feedback in the control of rhythmic movements,” Motor Control, Vol. 6, No. 2, pp. 129-145, 2002.

[3] N. Ayache and O. Faugeras, “Building, registrating, and fusing noisy visual maps,” International Journal of Robotics Research, Vol. 7, No. 6, pp. 45-65, 1988.

[4] J. L. Crowley, “World modeling and position estimation for a mobile robot using ultrasonic ranging,” IEEE International Conference on Robotics and Automation, Vol. 2, pp. 674-680, 1989.

[5] R. Chatila and J. Laumond, “Position referencing and consistent world modeling for mobile robots,” IEEE International Conference on Robotics and Automation, Vol. 2, pp. 138-145, 1985.