DocumentCode :
1619180
Title :
Elevator group control algorithm based on residual gradient and Q-learning
Author :
Zong, Ziliang ; Wang, Xugang ; Tang, Zheng ; Zeng, Guangzhou
Author_Institution :
Fac. of Comput. Sci., Shandong Univ., Jinan, China
Volume :
1
fYear :
2004
Firstpage :
329
Abstract :
This paper describes the application of a novel reinforcement learning to the difficult real world problem of elevator dispatching. We propose a new algorithm combing Q-learning and residual gradient to solve this problem and obtain the results which are better than other traditional elevator control algorithms.
Keywords :
dispatching; gradient methods; learning (artificial intelligence); lifts; Q-learning; elevator dispatching; elevator group control algorithm; reinforcement learning; residual gradient;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
SICE 2004 Annual Conference
Conference_Location :
Sapporo
Print_ISBN :
4-907764-22-7
Type :
conf
Filename :
1491420
Link To Document :
بازگشت