Linear Hopfield networks and constrained optimization

It is shown that a Hopfield neural network (with linear transfer functions) augmented by an additional feedforward layer can be used to compute the Moore-Penrose generalized inverse of a matrix. The resultant augmented linear Hopfield network can be used to solve an arbitrary set of linear equations or, alternatively, to solve a constrained least squares optimization problem. Applications in signal processing and robotics are considered. In the former case the augmented linear Hopfield network is used to estimate the "structured noise" component of a signal and adjust the parameters of an appropriate filter on-line, whereas in the latter case it is used to implement an on-line solution to the inverse kinematics problem.

[1]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[2]  Hua Li,et al.  Solving Simultaneous Linear Equations Using Recurrent Neural Networks , 1994, Inf. Sci..

[3]  John J. Hopfield,et al.  Simple 'neural' optimization networks: An A/D converter, signal decision circuit, and a linear programming circuit , 1986 .

[4]  J. Guo,et al.  A solution to the inverse kinematic problem in robotics using neural network processing , 1989, International 1989 Joint Conference on Neural Networks.

[5]  J. Hopfield,et al.  Computing with neural circuits: a model. , 1986, Science.

[6]  Bruno Siciliano,et al.  A solution algorithm to the inverse kinematic problem for redundant manipulators , 1988, IEEE J. Robotics Autom..

[7]  Karl Mathia,et al.  Solutions of linear equations and a class of nonlinear equations using recurrent neural networks , 1996 .

[8]  Robert M. Pap,et al.  Handbook of neural computing applications , 1990 .

[9]  M. Lothers,et al.  A neurocontroller for robotic applications , 1992, [Proceedings] 1992 IEEE International Conference on Systems, Man, and Cybernetics.

[10]  John Baillieul,et al.  Kinematic programming alternatives for redundant manipulators , 1985, Proceedings. 1985 IEEE International Conference on Robotics and Automation.

[11]  George G. Lendaris,et al.  Linear Hopfield networks, inverse kinematics and constrained optimization , 1994, Proceedings of IEEE International Conference on Systems, Man and Cybernetics.

[12]  George G. Lendaris,et al.  Hardware neural network implementation of tracking system , 1994, Proceedings of IEEE Workshop on Neural Networks for Signal Processing.

[13]  John J. Craig,et al.  Introduction to Robotics Mechanics and Control , 1986 .

[14]  R.N. Scott,et al.  Myoelectric signal analysis using neural networks , 1990, IEEE Engineering in Medicine and Biology Magazine.

[15]  John J. Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities , 1999 .

[16]  K. S. Banerjee.  Generalized Inverse of Matrices and Its Applications , 1973 .

[17]  Charles A. Klein,et al.  Review of pseudoinverse control for use with kinematically redundant manipulators , 1983, IEEE Transactions on Systems, Man, and Cybernetics.

[18]  Simon Haykin,et al.  Classification of radar clutter in an air traffic control environment , 1991, Proc. IEEE.

[19]  Yasmina Bestaoui,et al.  An unconstrained optimization approach to the resolution of the inverse kinematic problem of redundant and nonredundant robot manipulators , 1991, Robotics Auton. Syst..

[20]  John Edwards,et al.  Adaptive semi-autonomous robotic neurocontroller , 1994, Defense, Security, and Sensing.