基于图像处理的低误差机械臂位姿检测仿真

Simulation of pose detection of low error manipulator based on image processing

传统柔性机械臂位姿检测方法在实际应用中无法准确检测机械臂位资的俯仰角、偏航角以及滚转角。为此,提出基于机械视觉的柔性机械臂位姿检测方法。利用TL5147对机械臂位姿图像采集并显示。,通过EDMA传输法完成图像传输,提升程序工作效率。利用四元数算法完成柔性机械臂的空间位姿关系检测,保证机械臂位姿检测的准确性,实现机械视觉的柔性机械臂位姿检测。实验结果表明,通过对柔性机械臂位姿的俯仰角误差测试、偏航角误差测试以及滚转角误差测试,验证了所提柔性机械臂位姿检测方法的精准度高、可靠性强。

The traditional position and attitude detection method of flexible manipulator can not accurately detect the pitch angle,yaw angle and roll angle of manipulator in practical application.In this regard,this paper reports a pose detection method of flexible manipulator based on mechanical vision.Based on TL5147,the pose image of the manipulator was collected and displayed.According to the EDMA transmission method,the image transmission was completed for improving the efficiency of the program.The detection of the spatial pose relationship of the flexible manipulator was achieved via the quaternion algorithm,thus ensuring the accuracy of the pose detection of the manipulator.Eventually,the pose detection of the flexible manipulator based on mechanical vision was realized.The experimental results show that the detection method has high accuracy and excellent reliability.