智能空间下基于异质传感信息融合的目标跟踪  

Target tracking based on heterogeneous sensor information fusion in intelligent space

作　　者：桑森[1] 田国会[1] 段朋[1] 吴皓[1] 

机构地区：[1]山东大学控制科学与工程学院,济南250061

出　　处：《计算机工程与应用》2014年第16期48-53,66,共7页Computer Engineering and Applications

基　　金：国家高技术研究发展计划(863)(No.2009AA04Z220);山东省自然科学基金资助课题(No.ZR2011FM011)

摘　　要：针对大范围未知环境下的机器人目标跟踪问题,在智能空间下分布式智能网络设备与机器人本体二维激光的基础上,提出了一种基于异质信息融合结构以实现机器人对目标实时检测和跟踪的方法。系统通过颜色信息进行移动目标匹配,根据三角测量原理基于最小二乘法对目标进行三维重建;检测出运动目标后,通过激光传感器扫描目标人腿进行近邻点聚类获得目标的准确深度信息;利用一种优化的迭代扩展卡尔曼滤波算法对异质传感器进行信息融合,以实现基于智能空间的机器人定位与目标跟踪。实验结果验证了方法的有效性。For a wide range of detection and tracking of target for a mobile service robot problem under the unknown environment, this article puts forward a method based on the heterogeneous information fusion structure to realize the target detection and tracking in real-time of robots, whose basis is distributed intelligent network devices and the robot’s 2D laser range finder in the intelligence space. The system matches the moving target by the information on color, and carries out the 3D reconstruction of the target depending on the triangulation technique and least square method. The moment it detects the moving target, in order to get the exact distance information, it will scan the target human’s leg by the laser range finder to cluster the nearest neighbor. Meanwhile, this article raises a better Iterative Extend Kalman Filter for the heterogeneous sensor information fusion to realize the simultaneous robot localization and target tracking based on the intelligence space. The experimental results verify the effectiveness of the proposed method.

关 键 词：目标跟踪 分布式智能网络设备 扩展卡尔曼滤波 异质传感器信息融合 

分 类 号：TP242.3[自动化与计算机技术—检测技术与自动化装置]