以模糊類神經網路為基礎之多軸機械臂適應性定位及無速度感測控制設計

Design of Adaptive Positioning and Velocity Sensorless Control for Multi-link Robot Manipulator Via Fuzzy Neural Network

楊志偉(Zhi-Wei Yang)

指導教授 ： 魏榮宗

元智大學/電機通訊學院/電機工程學系/碩士(2008年)

摘要

本論文主要目的在於發展以模糊類神經網路為基礎之多軸機械臂適應性定位及無速度感測控制設計，其包含間接型適應模糊類神經網路控制、無速度感測模糊類神經網路控制和直接型適應模糊類神經網路控制，以達高精確度之機械臂定位性能。一般而言，由於機械臂實際應用時存在摩擦力、外部干擾、參數變化、等之不確定因素，倘若欲設計無需系統參數及速度/加速度回授資訊之控制架構以達精密定位控制目標，確實是值得挑戰的課題。間接型適應模糊類神經網路控制中，構思一個具有即時學習能力的連續時間T-S模糊模型動態來表示多軸機械臂的系統動態；再者，設計模糊類神經網路估測器即時調整位於局部模糊模型動態的非線性函數，以發展穩定的間接型適應模糊類神經網路控制法則。無速度感測模糊類神經網路控制中，其包含非線性觀測器和模糊類神經網路控制器，設計過程中無需預先瞭解系統資訊；非線性觀測器用來觀測機械臂的速度資訊，並設計不需要額外輔助補償器之模糊類神經網路控制架構。直接型適應模糊類神經網路控制中，直接設計模糊類神經網路控制器去近似一個基於動態模型所預先決定的穩定控制法則，並且僅使用機械臂的位置資訊即可達到穩定的控制性能。藉由里亞普諾(Lyapunov)穩定理論推導出所提出的三個控制法則及其相對應的網路參數適應調整演算法，以確保穩定的控制性能，並藉由直流伺服馬達驅動之雙軸機械臂的數值模擬與實作結果佐證所提出控制系統之有效性與強健性。此外，亦與比例微分控制、基於模糊模型控制、T-S型式模糊類神經網路控制和強健型類神經模糊網路控制之先前控制策略作一性能比較，以顯示本論文所提出控制系統之優越性。

關鍵字

T-S模糊模型 ； 模糊類神經網路 ； 無速度感測器 ； 非線性觀測器 ； 適應控制 ； 里亞普諾(Lyapunov)穩定性分析 ； 機械臂

並列摘要

This thesis focuses on the development of indirect adaptive fuzzy-neural-network control (IAFNNC), fuzzy-neural-network velocity sensorless control (FNNVSC) and direct adaptive fuzzy-neural-network control (DAFNNC) schemes for an n-link robot manipulator to achieve high-precision position tracking. In general, it is difficult to design a control approach without the model parameters and the joint velocity/acceleration information to achieve this control objective due to the uncertainties in practical applications, such as friction forces, external disturbances and parameter variations. In order to cope with this problem, three control topologies are investigated without the requirement of prior system information. In the IAFNNC, a continuous-time Takagi-Sugeno (T-S) dynamic fuzzy model with on-line learning ability is constructed for representing the system dynamics of an n-link robot manipulator. Then, a fuzzy-neural-network (FNN) estimator is designed to tune the nonlinear dynamic function vector in fuzzy local models, and the estimative vector is used to indirectly develop a stable IAFNNC law. The FNNVSC scheme including a nonlinear observer and a FNN controller is investigated without the requirement of prior system information. This nonlinear observer is used to estimate joint velocities of the robot manipulator, and a four-layer FNN is utilized for the major control role without auxiliary compensated control. In the DAFNNC, a FNN controller is directly designed to imitate a predetermined model-based stabilizing control law, and then the stable control performance can be achieved by only using joint position information. All the IAFNNC, FNNVSC and DAFNNC laws and the corresponding adaptive tuning algorithms for FNN parameters are established in the sense of Lyapunov stability analyses to ensure the stable control performance. Numerical simulations and experimental results of a two-link robot manipulator actuated by dc servomotors are given to verify the effectiveness and robustness of the proposed methodologies. In addition, the superiority of the proposed control schemes is indicated in comparison with previous researches including proportional-differential control (PDC), fuzzy-model-based control (FMBC), T-S type fuzzy-neural-network control (T-FNNC), and robust-neural-fuzzy-network control (RNFNC).

並列關鍵字

T-S fuzzy model ； Fuzzy neural network ； Velocity sensorless ； Nonlinear observer ； Adaptive control ； Lyapunov stability analyses ； Robot manipulator

參考文獻

[1] J. Moreno, R. Kelly, and R. Campa, “Manipulator velocity control using friction compensation,” IEE Proc. Contr. Theory Appl., vol. 150, no. 2, pp. 119-126, March 2003.

[2] E. M. Jafarov, M. N. A. Parlakçı, and Y. Istefanopulos, “A new variable structure PID-controller design for robot manipulators,” IEEE Trans. Contr. Syst. Technol., vol. 13, no. 1, pp. 122-130, Jan. 2005.

[3] V. Santibanez, R. Kelly, and M. A. Llama, “A novel global asymptotic stable set-point fuzzy controller with bounded torques for robot manipulators,” IEEE Trans. Fuzzy Syst., vol. 13, no. 3, pp. 362-372, June 2005.

[4] W. Gao and R. R. Selmic, “Neural network control of a class of nonlinear systems with actuator saturation,” IEEE Trans. Neural Netw., vol. 17, no. 1, pp. 147-156, Jan. 2006.

[5] E. Kim, “Output feedback tracking control of robot manipulators with model uncertainty via adaptive fuzzy logic,” IEEE Trans. Fuzzy Syst., vol. 12, no. 3, pp. 368-378, June 2004.