使用模糊類神經網路控制於非線性時變延遲系統及其應用於以網路為基礎之輪型機器人的軌跡追蹤

FUZZY-NEURAL-BASED CONTROL FOR NONLINEAR TIME-VARYING DELAY SYSTEMS WITH APPLICATION TO INTERNET-BASED TRAJECTORY TRACKING OF A CAR-LIKE WHEELED ROBOT

張立瑞(Li-jui Chang)

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

本論文中，一個部份已知並具有輸入及狀態時變延遲的系統，利用固定延遲的狀態空間模型的N個線性次系統去近似此非線性系統，再以放射狀為準則的類神經網路去學習因為時變延遲、擾動及modeling error所造成的不確定因素，再以此學習模式來設計額外的補償，以達到最佳遠端即時控制，由於網路傳輸會有干擾及不確定因素產生，所以本論文將假設網路控制為一個非線性時間延遲系統。接著，以所設計的控制器應用於以網路為基礎的智慧型空間之輪型機器人在的折線軌跡追蹤並且閃避路徑中之靜、動態障礙物。
在智慧型空間中，輪型機器人上的微處理機用於傳送馬達之回授訊號給server 端電腦並且執行由server端電腦所傳來的控制訊號，而在sever端電腦連接兩部CCD 攝影機以取得輪型機器人在智慧型空間中的大地座標位置及方位，再將此訊息及馬達之角度傳至client 端電腦，進行軌跡之規劃及控制訊號之計算，再由同路徑回傳控制訊號。

關鍵字

模糊線性模型 ； 類神經網路 ； 非線性時變延遲系統 ； 學習法則 ； 軌跡追蹤 ； 智慧型空間 ； 輪型機器人

並列摘要

Abstract --- In this thesis, a partially known nonlinear dynamic system with input and state time-varying delays is approximated by N fuzzy-based linear subsystems described by state-space model with average-delay. The proposed control contains a radial basis neural network to learn the uncertainties caused by the fuzzy-model error (e.g., time-varying delay, parameter variations) and the interactions resulting from the other subsystems. As the norm of the switching surface is inside of a defined set, the learning law starts; in this situation, the proposed method is an adaptive control possessing a compensation of uncertainties. No assumption about the upper bound of the time-varying delay for the state and the input is required. However, a time-average delay is needed for simplifying the controller design, and the stabilized conditions for every transformed delay-free subsystem must be satisfied. The stability of the overall system is verified by Lyapunov stability theory. Simulations as compared with a linear state feedback with integration control are also arranged to consolidate the usefulness of the proposed control.
To implement trajectory tracking and obstacle avoidance, two distributed CCD (charge-coupled device) cameras are set up to capture the dynamic position of the wheeled robot and the obstacle in a network-based intelligent space. Based on the control authority of these two CCD cameras, a suitable reference command including desired steering angle and forward-backward velocity for the FNBC in the client computer (or central controller) is planned. Finally, a sequence of experiments including the control of unloaded CLWR and the trajectory tracking of CLWR is carried out to consolidate the usefulness of the proposed control system.

並列關鍵字

Internet-based intelligent space ； Trajectory tracking ； Learning law ； Nonlinear time-varying delay system ； Fuzzy linear model ； Wheeled robot ； Radial-basis function

參考文獻

[1] R. J. Anderson and M. W. Spong, “Bilateral control of teleoperators with time delay,” IEEE Trans. Automat. Contr., vol. 34, no. 5, pp. 494-501, 1989.

[2] S. Mascolo, “Smith’s principle for congestion control in high-speed data networks,” IEEE Trans. Automat. Contr., vol. 45, no. 2, pp. 358-364, 2000.

[3] S. Munir and W. J. Book, “Internet-based teleoperation using wave variables with prediction,” IEEE/ASME Trans. Mechatron., vol. 7, no. 2, pp.124-133, 2002.

[4] F. Mazenc, S. Mondi and S. I. Niculescu, “Global asymptotic stabilization for chains of integrators with a delay in the input,” IEEE Trans. Automat. Contr., vol. 48, no. 1, pp. 57-63, 2003.

[5] Y. Tipsuwan and M.-Y. Chow, “Control methodologies in networked control systems,” Control Engineering Practice, vol. 11, pp. 1099-1111, 2003.