Abstract
Zhang neural networks (ZNN), a special kind of recurrent neural networks (RNN) with implicit dynamics, have recently been introduced to generalize to the solution of online time-varying problems. In comparison with conventional gradient-based neural networks, such RNN models are elegantly designed by defining matrix-valued indefinite error functions. In this paper, we generalize, investigate and analyze ZNN models for online time-varying full-rank matrix Moore–Penrose inversion. The computer-simulation results and application to inverse kinematic control of redundant robot arms demonstrate the feasibility and effectiveness of ZNN models for online time-varying full-rank matrix Moore–Penrose inversion.
Similar content being viewed by others
References
Klein CA, Kee KB (1989) The nature of drift in pseudoinverse control of kinematically redundant manipulators. IEEE Trans Robot Autom 5(2): 231–234
Yahagi T (1983) A deterministic approach to optimal linear digital equalizers. IEEE Trans Acoust Speech Signal Process 31(2): 491–500
Liu J, Chen S, Tan X, Zhang D (2007) Efficient pseudoinverse linear discriminant analysis and its nonlinear form for face recognition. Int J Pattern Recogn Artif Intell 21(8): 1265–1278
Hartmann WM, Hartwig RE (1996) Computing the Moore-Penrose inverse for the covariance matrix in constrained nonlinear estimation. SIAM J Optim 6(3): 727–747
Baushke HH, Borwein JM, Wang XF (2007) Fitzpatrick functions and continuous linear monotone operators. SIAM J Optim 18(3): 789–809
Dean P, Porrill J (1998) Pseudo-inverse control in biological systems: a learning mechanism for fixation stability. Neural Netw 11(7–8): 1205–1218
Bini DA, Codevico G (2003) Barel MV solving Toeplitz least squares problems by means of Newton’s iteration. Numer Algorithms 33(1–4): 93–103
Wei Y, Cai J, Ng MK (2004) Computing Moore-Penrose inverses of Toeplitz matrices by Newton’s iteration. Math Comput Model 40: 181–191
Chen L, Krishnamurthy EV, Macleod I (1994) Generalised matrix inversion and rank computation by successive matrix powering. Parallel Comput 20(3): 297–311
Zhou J, Zhu YM, Li XR, You ZS (2002) Variants of the Greville formula with applications to exact recursive least squares. SIAM J Matrix Anal Appl 24(1): 150–164
Ben-Israel A, Greville TNE (2003) Generalized inverses: theory and applications, 2nd edn. Springer-Verlag, New York
Wang H, Li J, Liu H (2006) Practical limitations of an algorithm for the singular value decomposition as applied to redundant manipulators. In: Proceedings of IEEE conference on robotics, automation and mechatronics, vol 1, pp 1–6
Zhang Y, Li Z, Fan Z, Wang G (2007) Matrix-inverse primal neural network with application to robotics. Dyn Contin Discrete Impuls Syst Ser A 14: 400–407
Wang J (1997) Recurrent neural networks for computing pseudoinverses of rank-deficient matrices. SIAM J Sci Comput 18(5): 1479–1493
Zhang Y, Ge SS (2005) Design and analysis of a general recurrent neural network model for time-varying matrix inversion. IEEE Trans Neural Netw 16(6): 1477–1490
Zhang Y, Wang J (2002) A dual neural network for convex quadratic programming subject to linear equality and inequality constraints. Phys Lett A 298(4): 271–278
Wang QF (2007) Theoretical and computational issues of optimal control for distributed Hopfield neural network equations with diffusion term. SIAM J Sci Comput 29(2): 890–911
Lingjærde OC, Liestøl K (1998) Generalized projection pursuit regression. SIAM J Sci Comput 20(3): 844–857
Smaoui N (2001) A model for the unstable manifold of the bursting behavior in the 2d Navier-Stokes flow. SIAM J Sci Comput 23(3): 824–840
Driessche PVD, Zou XF (1998) Global attractivity in delayed Hopfield neural network models. SIAM J Appl Math 58(6): 1878–1890
Hopfield JJ (1982) Neural networks and physical systems with emergent collective computational abilities. Proc Natl Acad Sci USA 79: 2554–2558
Hornik K, Stinchcombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Netw 2: 359–366
Yildiz N (2005) Layered feedforward neural network is relevant to empirical physical formula construction: a theoretical analysis and some simulation results. Phys Lett A 345(1–3): 69–87
Wei Y (2000) Recurrent neural networks for computing weighted Moore-Penrose inverse. Appl Math Comput 116(3): 279–287
Wei Y, Wu H, Wei J (2000) Successive matrix squaring algorithm for parallel computing the weighted generalized inverse \({A^+_{MN}}\) . Appl Math Comput 116(3): 289–296
Zhang Y, Jiang D, Wang J (2002) A recurrent neural network for solving Sylvester equation with time-varying coefficients. IEEE Trans Neural Netw 13(5): 1053–1063
Zhang Y, Li Z (2009) Zhang neural network for online solution of time-varying convex quadratic program subject to time-varying linear-equality constraints. Phys Lett A 373(18–19): 1639–1643
Zhang Y, Peng H (2007) Zhang neural network for linear time-varying equation solving and its robotic application. In: Proceedings of the 6th international conference on machine learning and cybernetics, vol 1, pp 3543–3548
Zhang Y, Fan Z, Li Z (2007) Zhang neural network for online solution of time-varying Sylvester equation. In: Proceedings of the 2nd international symposium on intelligent computation and application. Lecture notes on bioinformatics, pp 276–285
The MathWorks Inc (2007) Using Simulink, version 6.6, Natick, MA
Fill JA, Fishkind DE (1999) The Moore–Penrose generalized inverse for sums of matrices. SIAM J Matrix Anal Appl 21(2): 629–635
Parks-Gornet J, Imam IN (1989) Using rank factorization in calculating the Moore-Penrose generalized inverse. In: Proceedings of IEEE energy and information technologies in the southeast, vol 2, pp 427–431
Wiesel A, Eldar YC, Shamai S (2008) Zero-forcing precoding and generalized inverses. IEEE Trans Signal Process 56(9): 4409–4418
Ge SS, Lee TH, Harris CJ (1998) Adaptive neural network control of robotic manipulators. World Scientific, London
Zhang Y, Wang J (2002) Global exponential stability of recurrent neural networks for synthesizing linear feedback control systems via pole assignment. IEEE Trans Neural Netw 13(3): 633–644
Zhang Y, Chen K (2008) Global exponential convergence and stability of Wang neural network for solving online linear equations. Electron Lett 44(2): 145–146
Zhang Y (2005) Revisit the analog computer and gradient-based neural system for matrix inversion. In: Proceedings of IEEE international symposium on intelligent control, pp 1411–1416
Zhang Y, Ma W, Cai B (2009) From Zhang neural network to Newton iteration for matrix inversion. IEEE Trans Circuits-I 56(7): 1405–1415
Zhang Y, Cai B, Liang M, Ma W (2008) On the variable step-size of discrete-time Zhang neural network and Newton iteration for constant matrix inversion. In: Proceedings of the 2nd international symposium on intelligent information technology application, vol 1, pp 34–38
Zhang Y, Guo X, Ma W (2008) Modeling and simulation of Zhang neural network for online linear time-varying equations solving based on MATLAB Simulink. In: Proceedings of the 7th international conference on machine learning and cybernetics, pp 805–810
Zhang Y, Guo X, Ma W, Chen K, Cai B (2008) MATLAB Simulink modeling and simulation of Zhang neural network for online time-varying matrix inversion. In: Proceedings of IEEE international conference on networking, sensing and control, pp 1480–1485
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zhang, Y., Yang, Y., Tan, N. et al. Zhang neural network solving for time-varying full-rank matrix Moore–Penrose inverse. Computing 92, 97–121 (2011). https://doi.org/10.1007/s00607-010-0133-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00607-010-0133-9
Keywords
- Recurrent neural networks
- Time-varying problems
- Moore–Penrose inverse
- Inverse kinematic control
- Redundant robot arm