基于模糊路径纠偏的AGV视觉精定位研究

Research on precise positioning of AGV vision based on fuzzy path rectification

针对传统自主导航车(AGV)无法满足高精度路径跟踪和关键点精确定位问题,首次提出一种基于模糊控制的路径纠偏和二维码的关键点精确定位的组合方法。首先建立AGV四轮差动运动学模型,以位置偏差和角度偏差作为输入,纠偏控制量为输出设计一个二维模糊控制器实现快速纠偏。另外针对特殊点位高精度定位需求,根据每帧采样图片的尺寸固定这一特点建立基准坐标模型,通过三点定位法获取二维码轮廓,然后利用中心点的偏移计算误差,实现精确定位。

For the traditional autonomous navigation vehicle(AGV),the problem of high-precision path tracking and precise positioning of key points can not be satisfied.A combination method of key point precise location and path correction based on fuzzy control is proposed for the first time.Firstly,the four wheel differential kinematic model of AGV is established,position bias and angle deviation are used as inputs and the deviation rectification control is output.A two-dimensional fuzzy controller is designed to realize quick deviation correction.In addition,aiming at high-precision positioning requirements for special points,according to the size of each frame sampled image,the coordinate model is fixed.The two-dimensional code contour is obtained by three point positioning method,and then the offset is computed by the offset of the center point.Lastly,the precise positioning is got.