以RGB-D感測器實現全方向移動機器人自動對位控制

Design and Implementation of Alignment Control Based on RGB-D Sensors for Omni-directional Mobile Robot

楊晨佑(Chen-Yuo Yang)

指導教授 ： 練光祐

國立臺北科技大學/電資學院/電機工程系研究所/碩士(2013年)

https://doi.org/10.6841/NTUT.2013.00306

摘要

在機器人感知技術的發展中，機器人視覺之研究及其與運動控制系統之整合已獲得相當豐碩之成果。而使用視覺追蹤快速移動物體之研究則更具挑戰性，因為物體的快速移動，使系統反應時間變短，對影像處理與運動控制系統設計而言，都是嚴峻的挑戰。本論文所設計實現的系統，具有對深度影像辨識處理能力，可針對攝取的深度影像，做物件深度辨識動作，並且能根據物件的深度資訊轉換成以RGB-D 感測器為原點的三維空間位置，經由擷取多個物件位置後，可預測出物件的飛行軌跡與落地位置。本系統除了可預測出物件落地位置之外，還可驅動全方向移動機器人至預測的落地位置，同時成功的接住物件，並且都能即時完成。除此之外，根據不同物件，都可以進行上述功能。

關鍵字

RGB-D ； 影像追蹤 ； 自動對位控制

並列摘要

In the development of robotic sensing technology, the research of robot vision and its integration of motion control systems have gotten very fruitful achievement. The research of using robot visual to track a fast-moving object is a challenging task, because of the object with fast-moving shorten the reaction time of the system. The challenge comes from image processing and the design of motion control systems. The system in this thesis has the ability which can not only recognize the depth image but also change the depth information of object into 3-D position in terms of the coordinate frame of RGB-D sensors. After capturing several object positions, it can predict the flying path and landing position of the object. In addition to predicting the landing position of the object, this system also can drive the omni-directional mobile robot to the predicted landing position and successfully catch the object in time. All of these motions can complete in real time. Moreover, different sizes of objects can be carried out in the proposed scheme.

並列關鍵字

RGB-D Sensors ； Visual Tracking ； Alignment Control

參考文獻

[1] A. J. Lipton, H. Fujiyoshi and R. S. Patil, “Moving Target Classification and Tracking from Real-time Video,” IEEE Workshop on Applications of Computer Vision, pp. 8-14, 1998.

[2] D. S. Jang, G. Y. Kim and H. I. Choi, “Model-based Tracking of Moving Object,” Pattern Recognition, Vol. 30, No. 6, pp. 999-1008, 1997.

[7] D. Roller, K .Daniilidis and H.Nagel, "Model-based object tracking in monocular image sequences of road traffic scenes," International Journal of Computer Vision, pp. 257-281, 1993.

[8] Q. Delamarre and O. Faugeras, "3D Articulated Models and Multi-ViewTracking with Silhouettes," in Proc. of the Seventh IEEE International Conference on Computer Vision, Volume 2 , pp. 716 - 721, 20-27 Sept. 1999.

[9] I. Karaulova, P. Hall and A. Marshall ,"A hierarchical model of dynamics for tracking people with a single video camera," in Proc. of British Machine Vision Conference, pp. 262-352, 2000.