工业机器人末端执行器位置参数（TCP）是机器人离线编程及机器人末端工具误差校正的基础，研究快速、准确的TCP校准方法对保证工业现场环境下机器人系统顺利正常工作至关重要。以双目视觉测量为基础，结合空间坐标变换理论与机器人运动学，提出了一种应用于工业制造现场的机器人TCP参数快速自动校准方法。此方法具有非接触测量、校准速度快、精度高等优点，减少了传统接触式TCP校准过程中误差因素，并且克服了其标定速度慢，标定精度不足等缺点。结合ABB工业机器人对该方法进行验证实验，实验结果表明：对于直径为10 mm的末端工具，提出方法校准精度相对于传统接触式标定有很大的提高，可以满足工业现场高精度的、快速的TCP校准要求。

The TCP of robot is the fundation of off-line programming and end error correction. Studying fast and accurate calibration method is vital to ensure robots work normally in industrial field environments. In this paper, a TCP automatic calibration method was presented based on binocular vision measurement,combined with the spatial coordinate transformation theory and robot kinematics, which is applicable to industrial manufacturing site. This method has several advantages of non-contact, fast and high precision, which can reduce errors coming from previous calibrating method of contactless measurement and overcome its shortcoming of slowness. This method was testified on ABB robot and then experiment result shows that the accuracy is much better than that of conventional method when the calibration tool having a diameter of 10 mm. It can meet high precision, fast TCP calibration requirements in industrial site.