Application of Reinforcement Learning in Autonomous Navigation for Virtual Vehicles

被引：1

作者：

Niu, Lianqiang ^{[1
]}

Li, Ling ^{[2
]}

机构：

[1] Shenyang Univ Technol, Software Sch, Shenyang, Liaoning Prov, Peoples R China

[2] Shenyang Univ Technol, Informat Sci & Engn Sch, Shenyang, Liaoning Prov, Peoples R China

来源：

HIS 2009: 2009 NINTH INTERNATIONAL CONFERENCE ON HYBRID INTELLIGENT SYSTEMS, VOL 2, PROCEEDINGS | 2009年

关键词：

reinforcement learning; Q-learning; neural network; virtual vehicle; autonomous navigation;

D O I：

10.1109/HIS.2009.118

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

To solve the basic problem in the autonomous navigation for the virtual vehicles, the application of reinforcement learning is discussed, combined with artificial neural network, a model is proposed to realize the autonomous navigation for virtual vehicles. The autonomous navigation for the virtual vehicles is accomplished by reinforcement learning, the generalization problem for the action selection is solved by neural network, the model is tested by virtual environment, and the experiments show that the model is available and feasible.

引用

页码：30 / +

页数：2

共 7 条

[1] Reinforcement learning: A survey [J].

Kaelbling, LP ;

Littman, ML ;

Moore, AW .

JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 1996, 4 :237-285

[2]

MENG W, 2002, APPL RES COMPUTERS, V6, P79

[3]

Qin Zheng, 2007, Computer Engineering and Applications, V43, P215

[4]

Shi Mei-ping, 2004, Journal of System Simulation, V16, P1721

[5]

Sun Wei, 2006, Journal of PLA University of Science and Technology (Natural Science Edition), V7, P256

[6]

TOUZET CF, 1997, ROBOTICS AUTONOMOUS

[7]

WATKINS CJCH, 1992, MACH LEARN, V8, P279, DOI 10.1007/BF00992698

← 1 →