Title :
A neuromorphic controller with a human teacher
Author :
Guez, Allon ; Selinsky, John
Author_Institution :
Dept. of Electron. Comput. Eng., Drexel Univ., Philadelphia, PA, USA
Abstract :
Trainable adaptive controllers (TACs) are a subset of process controllers in which much of the design is done online by means of training rather than programming. The authors show how a neural-network-based architecture may be used to implement a general-purpose TAC. An example of controlling a cart-pole system (an inverted pendulum mounted on a cart) is provided. It is found that filtering of the human-teacher training data, using a dynamic model of the teacher, significantly improves the neuromorphic TAC´s performance.<>
Keywords :
adaptive control; computer architecture; computerised control; learning systems; neural nets; pendulums; cart-pole system; inverted pendulum; neural-network-based architecture; neuromorphic controller; online training; process controllers; trainable adaptive controllers; Adaptive control; Computer architecture; Digital control; Learning systems; Neural networks;
Conference_Titel :
Neural Networks, 1988., IEEE International Conference on
Conference_Location :
San Diego, CA, USA
DOI :
10.1109/ICNN.1988.23976
