An analysis of feature-based and state-based representations for module-based learning in mobile robots Export

@proceedings{citeulike:1714426, abstract = {The information available to robots in real tasks is widely distributed both in time and space, requiring the agent to search for relevant information. In this paper, we implement a solution that uses qualitative and quantitative knowledge to turn robot tasks able to be treated by reinforcement learning (RL) algorithms. The steps of this procedure include: 1) to decompose the overall task into smaller ones, using abstraction and macro-operators, thus achieving a discrete action space; 2) to use observation functions of the environment - here called features - to achieve both time and state space discretisation; 3) to use quantitative knowledge to design controllers that are able to solve the subtasks; 4) to learn the coordination of these behaviours using RL, more specifically Q-learning. The approach was verified on an increasingly complex set of robot tasks using a Khepera robot simulator. Two approaches for space discretisation were used, one based on features and the other on states. The learned policies over these two models were compared to a predefined hand-crafted one. It was found that the learned policy over the state-based discretisation leads quickly to good results, although it can not be applied to complex tasks, where the state space representation becomes computationally unfeasible.}, author = {Colombini, E. L. and Ribeiro, C. H. C.}, booktitle = {Hybrid Intelligent Systems, 2005. HIS '05. Fifth International Conference on}, citeulike-article-id = {1714426}, citeulike-linkout-0 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1587743}, journal = {Hybrid Intelligent Systems, 2005. HIS '05. Fifth International Conference on}, keywords = {abstraction, behavior, continuous\_time, learning, real\_world}, pages = {6 pp.+}, posted-at = {2007-10-01 09:40:49}, priority = {0}, title = {An analysis of feature-based and state-based representations for module-based learning in mobile robots}, url = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1587743}, year = {2005} }

TY - CONF ID - citeulike:1714426 L3 - citeulike-article-id:1714426 N2 - The information available to robots in real tasks is widely distributed both in time and space, requiring the agent to search for relevant information. In this paper, we implement a solution that uses qualitative and quantitative knowledge to turn robot tasks able to be treated by reinforcement learning (RL) algorithms. The steps of this procedure include: 1) to decompose the overall task into smaller ones, using abstraction and macro-operators, thus achieving a discrete action space; 2) to use observation functions of the environment - here called features - to achieve both time and state space discretisation; 3) to use quantitative knowledge to design controllers that are able to solve the subtasks; 4) to learn the coordination of these behaviours using RL, more specifically Q-learning. The approach was verified on an increasingly complex set of robot tasks using a Khepera robot simulator. Two approaches for space discretisation were used, one based on features and the other on states. The learned policies over these two models were compared to a predefined hand-crafted one. It was found that the learned policy over the state-based discretisation leads quickly to good results, although it can not be applied to complex tasks, where the state space representation becomes computationally unfeasible. JF - Hybrid Intelligent Systems, 2005. HIS '05. Fifth International Conference on TI - An analysis of feature-based and state-based representations for module-based learning in mobile robots T2 - Hybrid Intelligent Systems, 2005. HIS '05. Fifth International Conference on SP - 6 pp. KW - abstraction KW - behavior KW - continuous_time KW - learning KW - real_world AU - Colombini, EL AU - Ribeiro, CHC PY - 2005/// UR - http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1587743 ER -