论文部分内容阅读
针对传统的机器人柔性坐标测量方法中,机器人模型不完善及机器人固定参数不断变化导致测量精度难以提高的问题,提出一种基于双目视觉原理的全局实时校准方法,组建由两台相机组成的高精度全局校准单元,通过测量布置在机器人末端视觉传感器上的控制点阵,实时得到机器人末端的空间位姿,实现机器人在全局空间的精确定位。提出基于空间网格控制场的相机校准方法,构建像面坐标系上的残差库,实现相机在全视场空间内的高精度校准。实验表明,采用上述方法可实现±0.1mm的双相机校准精度,整个系统的测量精度可达±0.15mm,从根本上摆脱了机器人运动学模型及参数误差带来的影响,有效地保证了柔性坐标测量系统的精度。
Aiming at the problem that traditional robot’s flexible coordinate measurement method, the robot model is not perfect and the robot’s fixed parameters are changing constantly, it is difficult to improve the measurement accuracy. A global real-time calibration method based on binocular vision is proposed, The precision global calibration unit realizes the spatial pose of the robot in real time by measuring the control lattice arranged on the visual sensor at the end of the robot to realize the accurate positioning of the robot in the global space. A camera calibration method based on space grid control field is proposed to construct a residual library on the image plane coordinate system to achieve high precision camera calibration in the full field of view. Experiments show that the above method can be achieved ± 0.1mm dual camera calibration accuracy, the whole system measurement accuracy up to ± 0.15mm, fundamentally get rid of the robot kinematics model and parameter error of the impact effectively to ensure the flexibility Coordinate measurement system accuracy.