Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/8986
標題: 使用二階段學習實現簡化且精簡第二型模糊類神經網路 及其FPGA實現
A Simplified and Parsimonious Type-2 Fuzzy Neural Network with Two-Stage Learning and FPGA Implementation
作者: 張文昇
Jang, Wen-Sheng
關鍵字: Fuzzy chip
模糊晶片
fuzzy neural networks(FNNs)
structure learning
type-2 fuzzy systems
模糊類神經網路
架構學習
第二型模糊系統
出版社: 電機工程學系所
引用: [1] N. N. Karnik , J. M. Mendel, and Q. Liang, “Type-2 fuzzy logic systems,” IEEE Trans. on Fuzzy Systems, vol. 7, no. 6, pp. 643-658, 1999. [2] J. M. Mendel and R. I. John, “Type-2 fuzzy sets made simple,” IEEE Trans. On Fuzzy Systems, vol. 10, no. 2, pp. 117-127, 2002. [3] J. M. Mendel, Uncertain Rule-Based Fuzzy Logic System: Introduction and New Directions, Prentice Hall, Upper Saddle River, NJ2001. [4] Q. Liang and J. M. Mendel, “Equalization of nonlinear time-varying channels using type-2 fuzzy adaptive filters,” IEEE Trans. Fuzzy systems, vol. 8, no. 551-563, 2000. [5] H. B. Mitchell, “Pattern recognition using type-2 fuzzy sets,” Information Sciences, vol. 170, pp. 409-418, 2005. [6] P. Melin and O. Castillo, “Intelligent control of non-linear dynamic plants using type-2 fuzzy logic and neural networks,” Proc. IEEE Int. Conf. Fuzzy Systems, Budapest, Hungary, July, 2004. [7] H. Hagras, “A hierarchical type-2 fuzzy logic control architecture for autonomous mobile robots,” IEEE Trans. Fuzzy Systems, vol. 12, no. 524-539, 2004. [8] M. Melgarejo and C. Pena-Reyes, “Hardware architecture and FPGA implementation of a type-2 fuzzy system,” Proc. Of Great Lakes Symposium on VLSI (GLSVLSI), Boston, USA, pp. 458-261, 2004. [9] R. I. John, P. R. Innocent, and M. R. Barnes, “Neuro-fuzzy clustering of radiographic tibia image data using type-2 fuzzy sets,” Information Sciences, vol. 125, pp. 203-220, 2000. [10] A. G. Luigi Di Lascio and A. Nappi, “Medical differential diagnosis through type-2 fuzzy sets,” Proc. Of IEEE Int. Conf. Fuzzy Systems, pp. 371-376, 2005. [11] Q. Liang and J. M. Mendel, “Interval type-2 fuzzy logic systems: theory and design,” IEEE Trans. On Fuzzy Systems, vol. 8, no. 5, pp. 535-550, 2000. [12] C. H. Lee, Y. C. Lin, and W. Y. Lai, “Systems identification using type-2 fuzzy neural network (Type-2 FNN) systems,” Proc. IEEE Int. Symp. Computational Intelligence in Robotics and Automation, vol. 3, pp. 1264-1269, 2003. [13] C. H. Wang, C. S. Cheng, and T. T. Lee, “Dynamical optimal training for interval type-2 fuzzy neural network (T2FNN),” IEEE Trans. on Syst., , Man, and Cyber. - Part B: Cybernetics, vol. 34, no. 3, pp. 1462-1477, 2004. [14] J. M. Mendel, “Computing derivatives in interval type-2 fuzzy logic system,” IEEE Trans. On Fuzzy Systems, vol. 12. no. 1, pp. 84-98, Feb. 2004. [15] H. Hagras, “Comments on dynamical optimal training for interval type-2 fuzzy neural network (T2FNN),” IEEE Trans. Syst., Man and Cyber. - Part B: Cybernetics, vol. 36, no. 5, pp. 1206-1209, Oct. 2006. [16] C. F. Juang and Y. W. Tsao, “A self-evolving interval type-2 fuzzy neural network with on-line structure and parameter learning,” IEEE Trans. Fuzzy Systems, vol. 16, no. 6, pp. 1411-1424, Dec. 2008. [17] O. Uncu and . B. , “Discrete interval type-2 fuzzy system models using uncertainty in learning parameters,” IEEE Trans. Fuzzy Systems, vol. 15, no. 1, pp. 90-106, Feb. 2007. [18] C. F. Juang and C. T. Lin, “An on-line self-constructing neural fuzzy inference network and its applications,” IEEE Trans. Fuzzy Systems, vol. 6. no. 1, pp. 12-32, Feb. 1998. [19] C. F. Juang and Y. W. Tsao, “A type-2 self-organizing neural fuzzy system and its FPGA implementation”, IEEE Trans. Syst., Man, and Cyber., Part B: Cyber. vol.38, no.6, pp.1537-1548, 2008. [20] C. J. Lin and C. T. Lin, “An ART-based fuzzy adaptive learning control network,” IEEE Trans. Fuzzy Systems, vol. 5, no. 4, pp. 477-496, Nov. 1997. [21] M. Sugeno and T. Yasukawa, “A fuzzy logic based approach to qualitative modeling,” IEEE Trans. Fuzzy Syst., vol. 1, no. 1, pp. 7-31, Feb.1993. [22] H. Nakanishi, I. B. Turks﹐en, and M. Sugeno, “Areviewand comparison of six reasoning methods,” Fuzzy Sets Syst., vol. 57, pp. 257-295, 1993. [23] D. Kim and C. Kim, “Forecasting time series with genetic fuzzy predictor ensemble,” IEEE Trans. Fuzzy Systems, vol. 5, no. 4, pp. 523-535, 1997. [24] M. Russo, “Genetic fuzzy learning,” IEEE Trans. Evolutionary Computation, vol. 4, pp. 259-273, 2000. [25] K. B. Cho and B. H. Wang, “Radial basis function based adaptive fuzzy systems and their applications to system identification,” Fuzzy Sets Syst., vol. 83, pp. 325-339, 1996. [26] J. Kim and N. K. Kasabov, “HyFIS: Adaptive neuro-fuzzy inference systems and their application to nonlinear dynamic systems,” Neural Netw.,vol. 12, pp. 1301-1319, 1999. [27] D.Nauk andR.Kruse, “Neuro-fuzzy systems for function approximation,”Fuzzy Sets Syst., vol. 101, no. 2, pp. 261-271, 1999. [28] S.Wu and M. J. Er, “Dynamic fuzzy neural networks—A novel approach to function approximation,” IEEE Trans. Syst., Man, Cybern. B, Cybern.,vol. 30, no. 2, pp. 358-364, Apr. 2000. [29] M. Russo, “Genetic fuzzy learning,” IEEE Trans. Evol. Comput.,vol.4,no. 3, pp. 259-273, Sep. 2000. [30] S. Paul and S. Kumar, “Subsethood-product fuzzy neural inference system(SuPFuNIS),” IEEE Trans. Neural Netw., vol. 13, no. 3, pp. 578-599,May 2002. [31] Y. Gao and M. J. Er, “NARMAX time series model prediction: Feedfor-ward and recurrent fuzzy neural approaches,” Fuzzy Sets Syst., vol. 150,pp. 331-350, 2005. [32] A. S. Weigend and N. A. Gersehnfield, Time Series Prediction: Forecasting the Future and Understanding the Past, Reading, MA: Addison-Wesley, 1994. [33] S. Singh, “Noise impact on time-series forecasting using an intelligent pattern matching technique,” Pattern Recognition, vol. 32, no. 8, pp. 1389-1398, Aug. 1999. [34] J. L. Elman, “Finding structure in time ”, Cognit. Sci., vol.14, pp. 179-211 ,1990. [35] R. M. Neal, “Bayesian learning for neural networks,” Ph.D. dissertation,Dept. Comput. Sci., Univ. Toronto, Toronto, ON, Canada, 1995. [36] Jianke Zhu,StevenC.H.Hoi, and Michael Rung-Tsong Lyu, “Robust Regularized Kernel Regression,” IEEE Trans. Syst., Man, and Cyber., Part B: Cyber. vol.38, no.6, pp.1639-1643, Dec, 2008. [37] W. Chu, S. S. Keerthi, and C. J. Ong, “Bayesian support vector regression using a unified loss function,” IEEE Trans. Neural Netw., vol. 15, no. 1,pp. 29-44, Jan. 2004.
摘要: 本論文提出一個使用二階段學習實現簡化且精簡第二型模糊類神經網路(SPT2FNN)。SPT2FNN中的每一條模糊規則的前件部都是第二類型的模糊集合,而後件部使用Takagi-Sugeno-Kang (TSK)模糊規則形式。SPT2FNN使用簡化的擴展輸出計算,節省了電腦運算時間和硬體上面的成本。初始的規則集合一開始是空的。SPT2FNN使用二階段學習演算法,其建立是從第一型模糊規則延伸成第二型模糊規則。在第一階段的目標是透過即時的架構與參數學習來建立第一型模糊規則。第二階段首先擴展已建構好的第一型模糊規則成第二型模糊規則。此階段接下來調整後件部與前件部參數分別相對應使用卡門濾波器演算法和梯度下降法。SPT2FNN被應用在模擬系統辨識、股票價錢預測、渾沌訊號預測、實時序列預測、機器人手臂映射。在這些例子中和其他的第一型與第二型模糊系統互相比較,驗證SPT2FNN的成效與效率。對於SPT2FNN利用可程式邏輯閘陣列(FPGA)晶片實現上提出一個新的硬體實現技術。此簡化的函數運算可減少在硬體實現的成本。
This paper proposes a Simplified and Parsimonious Type-2 Fuzzy Neural Network with two-stage learning (SPT2FNN). The antecedent part in each fuzzy rule of SPT2FNN uses interval type-2 fuzzy sets and the consequent part is Takagi-Sugeno-Kang (TSK) type. The SPT2FNN uses a simplified extended-output-calculation operation to reduce the computation time and hardware implementation cost. The initial rule set in the SPT2FNN is empty. The SPT2FNN uses a two-stage learning algorithm to construct interval type-2 fuzzy rules from extension of type-1 fuzzy rules. The objective of the first stage is to construct type-1 fuzzy rules via online structure learning and parameter learning. The second stage first extends the constructed type-1 fuzzy rules to interval type-2 fuzzy rules, where highly overlapped type-1 fuzzy sets are merged to interval type-2 fuzzy sets to reduce the total number of fuzzy sets. This stage then tunes consequent and antecedent parameters in the type-2 fuzzy rules using rule-ordered Kalman filter algorithm and gradient descent algorithm, respectively. SPT2FNN has been applied to simulations on system identification, stock price prediction, chaotic signal prediction, real-time series prediction and the robot arm mapping problems. Comparisons with several type-1 and type-2 fuzzy systems in these examples have verified the effectiveness and efficiency of SPT2FNN. A new hardware circuit is proposed to implement the learned SPT2FNN in an FPGA chip. The simplified function in the SPT2FNN helps to reduce hardware implementation cost.
URI: http://hdl.handle.net/11455/8986
其他識別: U0005-2801201114580700
文章連結: http://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0005-2801201114580700
Appears in Collections:電機工程學系所

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.