传感器网络中利用反演集合估计的机器人定位方法

发布时间：2018-08-10 19:33

【摘要】：针对大型协作环境中移动机器人的全局定位问题,提出根据机器人车载传感器、环境传感器以及其他机器人的实时数据估计移动机器人的位置。首先,提出的方法整合大量不同类型传感器;然后,考虑了测量值数量可变、通用测角测量、受容错约束的测量统计知识等约束条件,将非线性边界误差估计问题看做一种反演集合;最后,处理特定类型的异常值和不精确环境下的模型误差。完成了误差和异常值的处理,就基本上获得了定位图,解决了移动机器人的定位问题。提出的方法利用实物实验进行验证。测试区域装备有多个传感器、固定在墙顶部的摄像机以及位于机器人上的可见标记。实验结果表明,提出的方法在协作环境中具有明显优势,处理异常值更加可靠。
[Abstract]:In order to solve the global localization problem of mobile robot in large-scale cooperative environment, the position estimation of mobile robot is proposed based on the real-time data of sensors, environment sensors and other robots. First of all, the proposed method integrates a large number of different types of sensors, and then, the constraints such as variable number of measurement values, general angle measurement, statistical knowledge of measurement subject to fault tolerance constraints are considered. The problem of nonlinear boundary error estimation is considered as an inversion set. Finally, the model errors in a given type of outliers and imprecise environments are dealt with. After the error and outliers are processed, the location diagram is basically obtained and the localization problem of the mobile robot is solved. The proposed method is verified by physical experiments. The test area is equipped with multiple sensors, cameras fixed to the top of the wall and visible markings on the robot. The experimental results show that the proposed method has obvious advantages in cooperative environment and is more reliable to deal with outliers.
【作者单位】：郑州大学西亚斯国际学院;厦门大学软件学院;
【基金】：国家自然科学基金资助项目(60873159)
【分类号】：TP242;TP212.9

【相似文献】