Neural Q-Learning Based Mobile Robot Navigation

被引:0
作者
Yun, Soh Chin [1 ]
Parasuraman, S.
Ganapathy, V. [2 ]
Joe, Halim Kusuma
机构
[1] Monash Univ, Sch Engn, Sunway Campus, Jalan Lagoon Selatan 46150, Bandar Sunway, Malaysia
[2] Univ Malaya, Fac Engn, Dept Elect Engn, Kuala Lumpur 50603, Malaysia
来源
MATERIALS SCIENCE AND INFORMATION TECHNOLOGY, PTS 1-8 | 2012年 / 433-440卷
关键词
Reinforcement Learning (RL); Q-Learning; Artificial Neural Network (ANN); Neural Q-Learning; Team AmigoBot (TM) robot and MATLAB;
D O I
10.4028/www.scientific.net/AMR.433-440.721
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
This research is focused on the integration of multi-layer Artificial Neural Network (ANN) and Q-Learning to perform online learning control. In the first learning phase, the agent explores the unknown surroundings and gathers state-action information through the unsupervised Q-Learning algorithm. Second training process involves ANN which utilizes the state-action information gathered in the earlier phase of training samples. During final application of the controller, Q-Learning would be used as primary navigating tool whereas the trained Neural Network will be employed when approximation is needed. MATLAB simulation was developed to verify and the algorithm was validated in real-time using Team AmigoBot (TM) robot. The results obtained from both simulation and real world experiments are discussed.
引用
收藏
页码:721 / +
页数:3
相关论文
共 12 条
[1]  
[Anonymous], MATLAB NEURAL NETWOR
[2]  
[Anonymous], 1989, THESIS CAMBRIDGE U
[3]  
Fitzpatrick J. B., 1981, MODERN MATH
[4]   The evolution of robotics research - From industrial robotics to field and service robotics [J].
Garcia, Elena ;
Antonia Jimenez, Maria ;
Gonzalez De Santos, Pablo ;
Armada, Manuel .
IEEE ROBOTICS & AUTOMATION MAGAZINE, 2007, 14 (01) :90-103
[5]   A reinforcement learning with evolutionary state recruitment strategy for autonomous mobile robots control [J].
Kondo, T ;
Ito, K .
ROBOTICS AND AUTONOMOUS SYSTEMS, 2004, 46 (02) :111-124
[6]   Application of artificial neural network based on Q-learning for mobile robot path planning [J].
Li, Caihong ;
Zhang, Jingyuan ;
Li, Yibin .
2006 IEEE INTERNATIONAL CONFERENCE ON INFORMATION ACQUISITION, VOLS 1 AND 2, CONFERENCE PROCEEDINGS, 2006, :978-982
[7]  
Mobile Robots Inc, 2007, TEAM AMIGOBOTTM OP M
[8]  
Saffiotti A, 1997, PROCEEDINGS OF THE SIXTH IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS, VOLS I - III, P573, DOI 10.1109/FUZZY.1997.616430
[9]  
Smart WD, 2002, 2002 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION, VOLS I-IV, PROCEEDINGS, P3404, DOI 10.1109/ROBOT.2002.1014237
[10]  
Smith Percey Franklyn, 1904, ELEMENTS ANAL GEOMET