基于红外检测的迹线跟踪和分支识别技术的研究

Study of Tracing and Branch Identification Technology Based on Infrared Detect

迹线跟踪和分支识别技术是车辆自动引导、无人驾驶、迹线机器人和寻迹机等应用领域的关键技术.提出一种用等间距排列的一组线阵状传感器对迹线的绝对位置和迹线与跟踪装置的相对交越变化情况进行实时检测的方法来达到快速寻迹的目的.通过对迹线绝对位置的检测可以直接判断出跟踪装置与迹线的偏离程度.而通过对迹线与跟踪装置的相对交越变化情况,可利用回归分析中的最小二乘法快速地计算出跟踪装置与迹线之间的偏离速度.根据跟踪装置与迹线之间的实时偏离程度和偏离速度,就可以采用相应的控制算法实现自动跟踪.用线阵式传感器对被跟踪迹线进行跟踪的精度和反应速度与传感器敏感元的排列间距具有直接的关系.因此,在允许的情况下,线阵式传感器的敏感元应采取较小的点距,这对迹线的实时性检测是有利的.

Trace following and branch identification technology is the key technology for self-direction of vehicle, unmanned drive, trace robot and trace finder. In order to find trace quickly, this paper introduces a real-time method with a set of equidistant linear array sensor to detect relative cross state and absolute position between trace and tracker. Offset degree between trace and tracker can be directly determined by detecting absolute position of trace. While offset velocity can be calculated with least square method of regression analysis by detecting relative cross state. According to real-time offset degree and offset velocity between trace and tracker, corresponding control algorithm is adopted to realize automatic following. As arranging distance between sensing elements of array sensor directly affects tracking accuracy and reaction speed with this method, smaller dot array distance should be kept between sensing elements of array sensor if permitted, and is advantageous for real-time detection of trace.