Title :
The study on a new intelligent control model ICM-LG and its application to the inverted pendulum
Author_Institution :
Dept. of Comput. Sci. & Technol., North China Electr. Power Univ., Baoding
Abstract :
An intelligent control model ICM-LG which inherits the advantages of linear quadratic regulator (LQR), human-imitating intelligent control theory and genetic algorithm is proposed in this paper. ICM-LG is independent of the accuracy of the controlled systempsilas mathematical model, neednpsilat analyze physical properties and control rules of system in detail, and avoids the pre-configuring control parameters and secondary manual regulation. By optimizing parameters with the unequal weights, the control process can reflect that the different control sub-targets have different priorities, and that the control demand can be focused on rapidity or stability differently. The stable control experiments for the inverted pendulum and the compares to LQR and the human-imitating intelligent control method show the validity and advantage of ICM-LG.
Keywords :
genetic algorithms; intelligent control; linear quadratic control; nonlinear control systems; pendulums; stability; ICM-LG; genetic algorithm; human-imitating intelligent control theory; intelligent control model; inverted pendulum; linear quadratic regulator; mathematical model; parameters optimisation; stable control experiments; Control system synthesis; Control systems; Cybernetics; Genetic algorithms; Intelligent control; Machine learning; Mathematical model; Nonlinear control systems; Optimal control; Regulators; Genetic algorithm; Human-imitating intelligent control; ICM-LG; Linear quadratic regulator; The inverted pendulum;
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620813
