数字滤波器在移动机器人超声信息融合中的应用

Application of Digital Filter to Map Building for Mobile Robot

超声传感器作为一种典型的测距传感器已被广泛应用于移动机器人的导航、避障与地图创建中.但是由于超声波传感器存在波束角,其方向性差且易发生多次反射等缺点,地图创建往往不甚理想,因此建立合适的超声波传感器模型并有效处理数据是关键.在分析超声常用模型的基础上,采用了弧线超声波模型,并且将"数字滤波器"的概念移植到超声信息融合上,旨在提高超声方向的可靠性,从而为地图创建提供更加准确的障碍特征信息,实验证明该方法在简单或复杂环境下均可有效地改善环境地图的创建效果.

As a typical distance sensor, the ultrasonic sensor has been widely used in mobile robot navigation, obstacle avoidance and map building. But it has some disadvantages, such as there are beam angle, poor direction and prone to reflections by some obstacles. Therefore, the key is to establish of a suitable model for uhrasonic sensors and effectively process ultrasonic data. In the paper, we use arc models of simplicity ultrasonic information. And then the digit filter is presented to improve the direction and fuse ultrasonic data, that can provide reliable characteristic information about barriers. Experiments prove that the method can improve the reliability of environment map building whether simple or complicated environment.