CAJ | 학술논문

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。
침대대범위미지배경하적궤기인목표근종문제，재지능공간하분포식지능망락설비여궤기인본체이유격광적기출상，제출료일충기우이질신식융합결구이실현궤기인대목표실시검측화근종적방법。계통통과안색신식진행이동목표필배，근거삼각측량원리기우최소이승법대목표진행삼유중건；검측출운동목표후，통과격광전감기소묘목표인퇴진행근린점취류획득목표적준학심도신식；이용일충우화적질대확전잡이만려파산법대이질전감기진행신식융합，이실현기우지능공간적궤기인정위여목표근종。실험결과험증료방법적유효성。
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.