基于多传感器信息融合的机器人障碍物检测.PDFVIP

第43卷第8期 中国测试 Vol.43 No.8 2017年8月 CHINA MEASUREMENT TEST August，2017 doi院10.11857/j.issn.1674-5124.2017.08.017 基于多传感器信息融合的机器人障碍物检测 王中立袁 牛 颖 渊太原理工大学信息工程学院袁山西 太原 030024冤 摘 要院针对单一传感器在机器人避障过程中不能全面且准确定位障碍物的缺点袁提出基于多传感器信息融合的障 碍物检测方法遥第一阶段使用视觉传感器检测未知环境中的障碍物袁通过Zernike矩边缘检测方法提取障碍物图像边 缘袁然后采用Hough变换原理提取障碍物的直线特征袁获得障碍物大概位置曰第二阶段使用超声波传感器和红外传感 器对障碍物进行检测袁获得障碍物准确位置曰最后使用联合卡尔曼滤波对3种传感器获得的信息进行融合袁得出融合 后的障碍物位置信息遥 实验结果表明院该方法克服视觉传感器尧超声波传感器和红外传感器的局限性袁可以精确感知 机器人周围的未知环境信息袁并能够检测和定位机器人路径上的障碍物袁定位误差6cm袁满足机器人避障的实时性 和可靠性需求遥 关键词院障碍物检测曰信息融合曰联合卡尔曼滤波曰视觉传感器曰Zernike矩曰Hough变换 文献标志码院A 文章编号院1674-5124渊2017冤08-0080-06 Obstacle detection of robot based on multi-sensor information fusion WANG Zhongli袁 NIU Ying 渊College of Information Engineering袁Taiyuan University of Technology袁Taiyuan 030024袁China冤 Abstract: Aiming at the shortcoming that the single sensor could not locate the obstacle completely and accurately in the process of obstacle avoidance of the robot袁 an obstacle detection method based on multi-sensor information fusion was proposed. Firstly袁 vision sensor was used to detect the obstacle in unknown environment. The edge of the obstacle image was extracted by the Zernike moment edge detection method袁 then the Hough transform principle was used to extract the straight line feature of the obstacle袁 so as to obtain the approximate position of the obstacle. Secondly袁 ultrasonic sensor and infrared sensor were used to detect the obstacle to obtain the exact position of obstacles. Finally袁 the federated Kalman filter was used to fuse the information obtained by the three sensors to gain infor