輪型機器人之設計與實現：使用適應性步階迴歸小腦模型控制器

Design and Implementation of Two-Wheeled Robot: Using Adaptive Backstepping CMAC Controller

林俊賢(Chun-Hsien Lin)

指導教授 ： 邱智煇

元智大學/電機通訊學院/電機工程學系/碩士(2011年)

摘要

在本論文中，我們實現了一個以倒單擺為基礎之輪型機器人。為完成輪型機器人的動作控制之目標，我們設計了一主動式平衡定位控制器以及轉向控制器，以雙輪的移動距離、機身的傾斜角度與轉向角度當作控制變數，主動控制馬達的出力。在本論文中我們使用了兩種方法作主控制器－小腦模型控制器以及適應性步階迴歸小腦模型控制器。小腦模型控制器是一種類似類神經網路的智慧型控制器。不同於類神經網路，小腦模型控制器有較快的學習速度、及較快的收斂速度。然而，此雙輪機器人存在著許多不確定之因素及雜訊，會影響系統動作，為了解決此問題，我們使用步階迴歸控制的設計方法設計適應性步階迴歸小腦模型控制器，將位置誤差加入角度誤差做角度及位置控制，此外我們使用一PD控制器做轉向控制，為了確保系統之穩定性，限制PD控制器輸出大小並加入一強健控制器 來縮小因不確定的動態系統與外在干擾和近似誤差產生的影響，並藉由李亞普諾夫定理(Lyapunov theory)推導其學習速率且證明其誤差之收斂；最後，我們也將透過輪型機器人平台之模擬與實驗，比較小腦模型控制器與適應性步階迴歸小腦模型控制器的模擬與實驗結果，驗證適應性步階迴歸小腦模型控制器在平衡定位控制上相對於小腦模型控制器有較佳的成效。

關鍵字

雙輪機器人 ； 小腦模型控制器 ； 步階迴歸控制 ； PD控制器 ； H infinite控制 ； 李亞普諾夫定理

並列摘要

In this thesis, we design and implementat a two-wheeled robot control system. To complete the motion control objective of the robot, we designed a balancing controller and a steering controller to control the robot. Since the dynamic characteristics of the robot system are nonlinear and time varying, it is difficult to design a suitable controller to achieve high-precision control at all time. A robust cerebellar model articulation controller (CMAC) via the backstepping control technique is proposed. In the backstepping CMAC control system, an adaptive CMAC is used to mimic an ideal backstepping control law. Moreover, we used a PD controller to do steering control. To ensure the stability of the system, we limited the PD controller output and used a robust controller to reduce the impacts of system uncertainty, outside disturbance and approximation error. The adaptation laws of the control system are derived in the sense of Lyapunov stability analysis, so that the stability of the system can be guaranteed. Finally, by the simulation and experimentation of two wheeled robot, we compare the CMAC controller and adaptive backstepping CMAC controller, respectively. From the results, the adaptive backstepping CMAC controller can be demonstrated that it has better performance for control and balance.

並列關鍵字

Lyapunov theory ； PD controller ； two wheeled robot ； cerebellar model articulation controller ； backstepping control

參考文獻

[5]F. Tanaka, B. Fortenberry, K. Aisaka, and J. R.Movellan, “Plans for Developing Real-time Dance Interaction between QRIO and Toddlers in a Classroom Environment” Development and Learning, 2005. Proceedings. The 4th International Conference on pp. 142-147, 19-21 July, 2005.

[12]R. Mak and J. Walton, “The collaborative information portal and NASA's Mars Rover mission”, IEEE Internet Computing,Vol. 9, No. 1, pp. 20-26, Jan.-Feb. 2005.

[13]S. B. Kim, M. Spenko, S. Trujillo, B. Heyneman, V. Mattoli, and M. R. Cutkosky, “Whole body adhesion: hierarchical, directional and distributed control of adhesive forces for a climbing robot”, IEEE International Conference on Robotics and Automation, pp. 1268-1273, 2007.

[16]林永春「自平衡載人運輸車之系統設計、建模與控制研究」中興大學電機工程研究所，碩士論文，2007年（蔡清池指導）。

[17]S. C. Lin and C. C. Tsai, “Development of a Self-Balancing Human Transportation Vehicle for the Teaching of Feedback Control,” IEEE Trans. education, Vo1. 52, No. 1, pp. 157-168, 2008, Februaey 2009.