DocumentCode :
2758181
Title :
A Heuristic Reinforcement Learning for Robot Approaching Objects
Author :
Wang, B. ; Li, J.W. ; Liu, H.
Author_Institution :
Robot Res. Inst., Harbin Inst. of Technol.
fYear :
2006
fDate :
1-3 June 2006
Firstpage :
1
Lastpage :
5
Abstract :
Autonomous approaching objects for an arm-hand robot is a very difficult problem because the possible arm-hand configurations are numerous. In this paper, we propose a modified reinforcement learning algorithm for a multifingered hand approaching target objects. The proposed approach integrates the heuristic search information with the learning system, and solves the problem of how an arm-hand robot approaches objects before grasping. In addition, this method also overcomes the problem of time consuming of traditional reinforcement learning in the initial learning phase. The algorithm is applied to an arm-hand robot to approach objects before grasping, which can enable the robot to learn approaching skill by trial-and-error and plan its path by itself. The experimental results demonstrate the effectiveness of the proposed algorithm
Keywords :
control engineering computing; heuristic programming; intelligent robots; learning (artificial intelligence); manipulators; search problems; arm-hand robot; heuristic reinforcement learning; heuristic search; multifingered hand; Acceleration; Aerodynamics; Grasping; Heuristic algorithms; Learning systems; Mechatronics; Orbital robotics; Path planning; Robot sensing systems; Sensor systems; A* search; Grasping; Multifingered hand; Reinforce learning;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Robotics, Automation and Mechatronics, 2006 IEEE Conference on
Conference_Location :
Bangkok
Print_ISBN :
1-4244-0024-4
Electronic_ISBN :
1-4244-0025-2
Type :
conf
DOI :
10.1109/RAMECH.2006.252749
Filename :
4018865
Link To Document :
بازگشت