最小二乘法在焊接点监控系统中的应用

The application of least square algorithm in soldered joint monitor system

工业控制系统中,为了监视焊接机器手臂的工作情况,要求监视摄像头始终对准焊接头。而在实际环境中很难获取焊接头三维坐标和摄像头转动参数之间的精确对应关系。为了找到它们之间精确的对应关系,拟采用最小二乘拟合算法,获取焊接头三维坐标和摄像头转动参数之间的变换参数,并对拟合方法中初始值进行优化,优化后获得误差极小的变换参数,使得焊头的三维坐标精确定位摄像头的转动参数。实验数据表明,新方法使得二者之间误差极小,满足实际设计需求。

In the industrial control system,in order to monitor the work of welding robot＇s arm,it requires surveillance camera always aiming at soldered joint.But in real environment it＇s difficult tofind the exact relation of three-dimensional coordinate of soldered joint to the camera rotation parameters.By using least square fitting algorithm,this paper is to get the transformation parameters between three-dimensional coordinate of soldered joint and camera rotation parameters,and to optimize the initial value in the fitting method.After the optimization,it gets a transform parameter with minimum error,which makes the three-dimensional coordinate of soldered joint locate accurately at camera rotation parameters.The data in the test show that this new method makes the minimum error,and meets the actual design requirement.