In this paper, we propose a measurement method for dynamic angles in large-space based on machine vision and servo control. The azimuth and pitch angles are obtained in real time by identifying and tracking a cooperative target in a high-precision servo system with a camera. Furthermore, the distance between the target and the camera is calculated by means of a monocular measurement. Based on the measured angles and distance, the dynamic angle of the rotating object in test is derived according to the transformational relations. In order to decrease measurement errors, an error correction model is established by the connection between the measurement distance and observation errors caused by the camera, dynamic tracking and pose changing of the target plate. A manual displacement platform with high precision in two-dimensions is conducted to validate the effectiveness of the proposed approach. The experimental results show that, compared with the existing measurement methods, the proposed method enlarges the measurement range of large-space angles to 18.788 m, and the measurement errors of the azimuth channel and pitch channel are 0.018° and 0.019°, respectively, which achieve a high measurement accuracy and verify the validity of the method.