以分散式主動視覺網路系統應用於輪型機器人之軌跡追蹤及避障

Distributed Active-Vision Network-Space Approach for Trajectory Tracking and Obstacle Avoidance of a Car-Like Wheeled Robot

施欽元(Chin-Yan Shih)

指導教授 ： 黃志良

大同大學/工程學院/機械工程學系所/碩士(2006年)

摘要

中文摘要
在本篇論文，應用模糊可變結構分散控制在分散式主動視覺網路系統並使輪型機器人達到折線軌跡追蹤及避障。完成此研究機構包括，雙自由度CCD攝影機旋轉平台，如，主動式CCD1和CCD2(或ACCD1和ACCD2)，一台輪型機器人。本系統為多樣取樣速率。首先經由ACCD1或ACCD2將輪型機器人或障礙物影像經由無線模組傳回個人電腦端並處理計算出輪型機械人、ACCD1和ACCD2之參考訊號。使用多層感知器去建立影像平面和世界座標系統間之關係。此三個參考訊號經由個人電腦端之無線模組傳送至輪型機械人、ACCD1、ACCD2。分別使用TI的DSP TMS320LF2407來控制輪型機器上的兩顆DC馬達及ACCD上之兩顆步進馬達。這兩個主動式視覺系統和輪型機器人是藉由三個模糊可變結構分散控制，其中兩個在個人電腦端，另一個則在輪型機器人上之DSP上運算，最後應用一連串之實驗來驗證所提出的系統之強健性。

關鍵字

軌跡追蹤 ； 輪型機器人 ； 主動式視覺網路系統 ； 模糊可變結構分散式控制 ； 多層感知器

並列摘要

Abstract -- In this thesis, the trajectory tracking of line segments and the obstacle avoidance for a wheeled robot within distributed active-vision network-space via fuzzy variable structure decentralized control (FVSDC) is developed. To implement trajectory tracking and obstacle avoidance, two distributed wireless CCD (charge-coupled device) cameras individually driven by two stepping motors, i.e., active CCD1 and CCD2 (or ACCD1 and ACCD2), are set up to capture the dynamic position of the wheeled robot or obstacle. The proposed control system includes quad-processors with multiple sampling rates. One personal computer is first employed to receive the image of the wheeled robot and the obstacle from ACCD1 or ACCD2 by wireless transmitter, and then to plan three reference commands for the wheeled robot, ACCD1, and ACCD2. The image processing and calibration between the world coordinate and the image plane coordinate by using multilayer perceptrons are also established. These three reference commands are transmitted by a wireless device on personal computer and received by the other three wireless devices on the wheeled robot, ACCD1, and ACCD2. The second, third and fourth processors are individually used a DSP TMS320LF2407 of TI Co. to control two DC motors of wheeled robot, two stepping motors of ACCD1, and two stepping motors of ACCD2. These two active-vision systems and the wheeled robot are controlled by three fuzzy variable structure decentralized controls (FVSDCs), two of them are in the PC and the other is in the DSP of the wheeled robot. Finally, a sequence of experiments is carried out to consolidate the performance of the proposed control system.

並列關鍵字

Active vision system ； Fuzzy variable structure decentralized control ； Obstacle avoidance ； Trajectory tracking ； Wheeled robot ； Network space ； Multi-processor.

參考文獻

[1] B. Sinopoli, C. Sharp, L. Schenato, S. Schaffert and S. S. Sastry, “Distributed control applications within sensor networks,” Proceedings of the IEEE, vol. 91, no. 8, pp. 1235-1246, 2003.

[2] T. Yamaguchi, E. Sato and Y. Takama, “Intelligent space and human centered robotics,” IEEE Trans. Ind. Electron., vol. 50, no. 5, pp. 881-889, 2003.

[3] J. H. Lee and H. Hashimoto, “Controlling mobile robots in distributed intelligent sensor network,” IEEE Trans. Ind. Electron., vol. 50, no. 5, pp. 890-902, 2003.

[5] T. S. Barger, D. E. Brown and M. Alwan, “Health-status monitoring through analysis of behavioral patterns,” IEEE Trans. Syst. Man & Cyber., Part A, vol. 35, no. 1, pp. 22-27, 2005.

[6] F. Doctor, H. Hagras and V. Callaghan, “A fuzzy embedded agent-based approach for realizing ambient intelligence in intelligent inhabited environments,” IEEE Trans. Syst. Man & Cyber., Part A, vol. 35, no. 1, pp. 55-65, 2005.