基于三色球的单目相机三维空间定位算法研究

Research on Three-Dimensional Space Positio ning Algorithm for Monocular Based on Tricolor Sphere

为了实现无工装夹具条件下的机器人焊接,提出基于三色球的单目相机三维空间定位方法,利用球在任意角度成像的几何不变性,结合球的半径信息实现球的深度信息的提取,解决了传统单目相机无法实现三维空间定位的难题。定位算法引入颜色滤波简化三色球的识别,改进霍夫圆检测算法,提高了三色球的识别精度。研究了球的深度信息计算模型,并推导了相关公式,实现了单目相机下球的三维坐标计算。构建了基于3个标准球的焊接工件定位模型,通过工件上3个定位点设置的标准球实现焊接工件定位。基于三维微位移平台设计了定位精度检测实验平台,并开展了相关实验。实验表明,三维空间定位算法的定位误差在±0.24mm以内,满足焊接工件定位的精度要求。

In order to realize robot welding without fixtures,the three-dimensional space positioning method of monocular camera based on three-color balls was proposed to solve the problem that tradit ional monocular cameras cannot achieve three-dimensional spatial positioning in this paper,which uses the geom etric invariance of ball imaging at any angle and combines the radius information of the ball to extract the depth information of the ball.The color filtering was introduced in the positi oning algorithm to simplify the identificati on of three-color balls,the Hough circle detection algorithm was modified,and the identification accuracy of three-color b alls was improved.The calculation model of the depth information of the ball was researched,and the relevant formulas were derived to realize the calculation of the three-dimensional coordi nates of the ball under the monocular camera.The welding workpiece positioning model based on three standard balls was established,and the welding workpiece was positioned by the standard balls set at the three positioning points of the workpiece.Based on the three-dimensional micro-displacement platform,an experimental platform for positioni ng accuracy detection was designed,and relat ed experiments were carried out.The experimental results show that the positioning error of the three-dimensional sp ace positioning algorithm is within±0.24mm,which meets the positioning a ccuracy requirements of weld workpiece..