Residual activation neural network

Details Residual activation neural network Residual activation neural network

1992-06-10

1994-10-04

Smith, Jerry

Boots, shoes, and leggings

364148, 395 23, G06F 1500, G05B 1300

Patent

active

053532074

ABSTRACT:
A plant (72) is operable to receive control inputs c(t) and provide an output y(t). The plant (72) has associated therewith state variables s(t) that are not variable. A control network (74) is provided that accurately models the plant (72). The output of the control network (74) provides a predicted output which is combined with a desired output to generate an error. This error is back propagated through an inverse control network (76), which is the inverse of the control network (74) to generate a control error signal that is input to a distributed control system (73) to vary the control inputs to the plant (72) in order to change the output y(t) to meet the desired output. The control network (74) is comprised of a first network NET 1 that is operable to store a representation of the dependency of the control variables on the state variables. The predicted result is subtracted from the actual state variable input and stored as a residual in a residual layer (102). The output of the residual layer (102) is input to a hidden layer (108) which also receives the control inputs to generate a predicted output in an output layer (106). During back propagation of error, the residual values in the residual layer (102) are latched and only the control inputs allowed to vary.

REFERENCES:
patent: 4858147 (1989-08-01), Conwell
patent: 4912753 (1990-03-01), Evans, Jr.
patent: 4928484 (1990-05-01), Peczkowski
patent: 4972363 (1990-11-01), Nguyen et al.
patent: 4979126 (1990-12-01), Pao et al.
patent: 5023045 (1991-06-01), Watanabe et al.
patent: 5111531 (1992-05-01), Grayson et al.
patent: 5113483 (1992-05-01), Keeler et al.
"Neural Controllers" by Demetri Psaltis, Athanasios Sideris and Alan Yamamura, IEEE First International Conference on Neural Networks, Jun. 21-24, 1987, IV-551-558.
"Towards Effective Neuomorphic Controllers" by Julio Tanomaru and Cigar Aimed, ICON '91, pp. 1395-1400.
"Neural networks for nonlinear internal model control" by K. J. Hunt and D. Sbarbaro, IEEE Proceedings-D, vol. 138, No. 5, Sep. 1991, pp. 431-438.
"A Nonlinear Receding Horizon Controller Based on Connectionist Models" by D. Sbarbaro and K. J. Hunt, Proceedings of the 30th IEEE Conference on Decision and Control, 1991, vol. 1, pp. 172-173.
"Connectionist Approach to Non-linear Internal Model Control Using Gaussian Approximation" by K. J. Hunt and D. Sbarbaro, Proceedings of the 1991 American Control Conference, vol. 2, Jun. 27, 1991, pp. 1826-1827.

Affiliated with

Also associated with

Rating

Residual activation neural network does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Residual activation neural network, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Residual activation neural network will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-586076