The Research of Elevator Group Dynamic Scheduling During Up-peak Traffic Based on POMDP

Author

Zong Qun ; Sun Zheng-ya

Author_Institution

Acad. of Electr. Eng. & Autom., Tianjin Univ., China

fYear

2006

fDate

7-11 Aug. 2006

Firstpage

1705

Lastpage

1709

Abstract

Due to the existence of the uncertainty of the traffic flows in the problem of elevator group control scheduling, Markov decision processes can´t deal with the model of the elevator group control scheduling well. The model based on partially observable Markov decision processes is studied to solve this problem. With application of the feedforward neural network, which is integrated into the linear Q-learning to construct the whole algorithm for elevator group scheduling, the optimal or nearly optimal policies are obtained gradually. Compared with previous algorithms, this algorithm greatly improves the adaptability to the up-peak traffic flow.

Keywords

Markov processes; dynamic scheduling; feedforward neural nets; learning (artificial intelligence); traffic; elevator group control scheduling; elevator group dynamic scheduling; elevator group scheduling; feedforward neural network; linear Q-learning; partially observable Markov decision processes; reinforcement learning; traffic flows; up-peak traffic flow; Automation; Communication system traffic control; Dynamic scheduling; Elevators; Intelligent control; Learning; Scheduling algorithm; Sun; Traffic control; Uncertainty; POMDP; elevator group control scheduling; linear Q-learning; reinforcement learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference, 2006. CCC 2006. Chinese

Conference_Location

Harbin

Print_ISBN

7-81077-802-1

Type

conf

DOI

10.1109/CHICC.2006.280826

Filename

4060383