基于克里金预测的机器人源定位方法  

Robotic source location prediction based on Kriging method

作　　者：曹凯 陈阳泉[1,2] 高嵩 曹牧寒[2] 王坤 CAO Kai;CHEN Yang-quan;GAO Song;CAO Mu-han;WANG Kun(Mechanical and Electrical Engineering School,Xi’an Technological University,Xi’an 710021,China;Electronic Information Engineering School,Xi’an Technological University,Xi’an 710021,China)

机构地区：[1]西安工业大学机电工程学院,陕西西安710021 [2]西安工业大学电子信息工程学院,陕西西安710021

出　　处：《计算机工程与设计》2021年第11期3261-3270,共10页Computer Engineering and Design

基　　金：陕西省国际科技合作计划基金项目(2019KW-014);陕西省重点研发计划基金项目(2019GY-072);陕西省教育厅专项科研计划基金项目(17JK0369)。

摘　　要：传统的基于梯度的算法对机器人底层控制及性能要求高,限制了源定位工作的效率与成功率。为此结合统计学方法和场的分布特点,在简化机器人底层控制的基础上,设计一种场探索的方法。在基于ROS的机器人平台上实现,通过设计机器人的遍历搜索算法使机器人能够按照预先定义的路径运动,得到数据后利用克里金法对整个光源强度分布进行预测。在真实实验场景下,验证了场探索方法的有效性,实验结果表明,该算法的成功率更高、稳定性更好,并直观地预测出了场的分布等值线,便于进一步分析与研究。Traditional gradient-based algorithms have high requirements on the robot’s underlying control and performance,which limits the efficiency and success rate of source location.To this end,a statistical method and the distribution characteristics of the field were combined,on the basis of simplifying the bottom control of the robot,a field exploration method was designed.This method was implemented on a ROS robot platform.By designing the robot’s traversal search algorithm,the robot moved according to a predefined path.After obtaining the data,the Kriging method was used to predict the entire light source intensity distribution.In the real experimental scenario,the effectiveness of the field exploration method was verified.Experimental results show that the proposed algorithm has higher success rate and better stability,and intuitively predicts the field distribution contour,which is convenient for further analysis and study.

关 键 词：源定位 场探索 ROS机器人 克里金法 等值线 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]