%0 Journal Article
%T Line-grasping control of de-icing robot based on k-nearest neighbor reinforcement learning<br>基于k–最近邻分类增强学习的除冰机器人抓线控制
%A WEI Shu-ning
%A WANG Yao-nan
%A YIN Feng
%A YANG Yi-min
%A <br>魏书宁
%A 王耀南
%A 印峰
%A 杨易旻
%J 控制理论与应用
%D 2012
%I 
%X The flexible mechanical characteristic of power lines induces difficulties for line-grasping control for de-icing robots. To deal with this difficulty, we propose for de-icing robots a line-grasping control approach which combines the k-nearest neighbor (KNN) algorithm and the reinforcement-learning (RL). In the learning iteration, the state-perception mechanism of the KNN algorithm selects k-nearest states and weights; from k-weighted states, an optimal action is determined. By expressing a continuous state by k-nearest discrete states in this way, this approach effectively ensures the convergence for the computation and avoids the curse of dimensionality occurred in traditional continuous state-space generalization methods. Abilities of RL in perception and adaptation to the environment make the line-grasping control to tolerate possible errors in robot model, errors of robot arm attitudes and interferences from the environment. The design procedures are presented in details. Simulation results of line-grasping control based on this approach are given.
%K de-icing robot
%K k-nearest neighbor
%K reinforcement learning
%K curse of dimension<br>除冰机器人
%K k–最近邻分类算法
%K 增强学习
%K 维数灾难
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=8240383F08CE46C8B05036380D75B607&jid=970898A57DFC021F93AB51667BAED7F7&aid=1D1ECD0B64DA3640BFDAA390FFB9C685&yid=99E9153A83D4CB11&vid=771469D9D58C34FF&iid=E158A972A605785F&sid=47F7649551A37CFC&eid=DA4893B5F9885621&journal_id=1000-8152&journal_name=控制理论与应用&referenced_num=0&reference_num=0