DocumentCode :
3229785
Title :
Continuous reinforcement learning algorithm for skills learning in an autonomous mobile robot
Author :
Boada, Maria Jesus L ; Egido, Veronica ; Barber, Ramon ; Salichs, Miguel Angel
Author_Institution :
Mech. Dept., Carlos III Univ., Madrid, Spain
Volume :
4
fYear :
2002
fDate :
5-8 Nov. 2002
Firstpage :
2611
Abstract :
Learning endows a mobile robot with a higher flexibility and allows it to adapt to changes occurring in the environment or in its internal state in order to improve its results. Based on this idea, this paper presents a reinforcement learning algorithm which allows the robot to learn simple skills such as go to goal and contour following. In the proposed learning algorithm the robot receives a real continuous reinforcement signal. Thus, it is not necessary to estimate an expected reward. Most of the robotic applications work with continuous variables such as velocity, position, sensors readings etc. The presented reinforcement learning algorithm is able to manage continuous input and output spaces. Finally, the robot is capable of performing the complex skill called go to go avoiding obstacles from the sequencing of previously learnt skills.
Keywords :
collision avoidance; learning (artificial intelligence); mobile robots; robot vision; RWI-1321 mobile robot; autonomous mobile robot; continuous input space management; continuous output space management; continuous reinforcement learning algorithm; continuous reinforcement signal; contour following; go to go avoiding obstacles; go to goal; laser sensor; skills learning; vision system; Intelligent robots; Intelligent sensors; Legged locomotion; Mobile robots; Neural networks; Orbital robotics; Robot sensing systems; Robotics and automation; Supervised learning; Systems engineering and theory;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
IECON 02 [Industrial Electronics Society, IEEE 2002 28th Annual Conference of the]
Print_ISBN :
0-7803-7474-6
Type :
conf
DOI :
10.1109/IECON.2002.1182805
Filename :
1182805
Link To Document :
بازگشت