注意事项和建议

图像质量

保证标定对象的图像充分曝光且对焦是非常重要的。为确保标定对象处于 depth of field ，建议使用更大的 \(f\)-值，并用更长的曝光时间进行补偿。此外，应对所有图像使用相同的光圈设置，以避免因光圈大小不同而造成不必要的波动。还有一个要点是不要使用相似的视角去拍摄手眼标定的图像，而应该充分利用标定对象和机器人的所有六个自由度进行成像。

Zivid API

Zivid点云以毫米为单位。 Zivid API中手眼输出的位姿的平移部分也以mm为单位。因此，手眼标定输入的机器人位姿的平移部分也必须以mm为单位。

用于检测的基准标记

如果标定对象是 Zivid calibration board ，请确保基准/ArUco标记在所有机器人位姿都可以被检测到。否则，算法在尝试检测棋盘格时将会失败。

Eye-to-hand

在eye-to-hand标定过程中，标定对象安装在机器人末端执行器上并随机器人移动。它可以直接固定在工具法兰上或由夹具固定。安装的确切位置并不重要，因为不必知道标定对象和末端执行器之间的相对位姿。重要的是标定对象在运动过程中不会出现相对于工具法兰或夹具的位移，它必须被良好地固定住或被夹具紧紧地抓住。建议使用由刚性材料制成的安装支架以及标定板。保证标定对象在图像采集期间保持不动是非常重要的。建议在机器人和标定对象定位完成等待几秒钟，直到整个结构从运动后的振动中稳定下来。应通过控制机器人的加速度使其平稳地运动，以避免标定对象发生晃动或错位。

Eye-in-hand

在eye-in-hand标定过程中，标定对象是固定放置在机器人工作区内的，这个固定位置需要保证安装在机器人上的相机可以从不同的视角观测到它。 标定对象的确切位置并不重要，因为不必知道其相对于机器人基座的位姿。 但是，标定对象在标定期间应牢牢固定住，不得移动。

环境条件

温度变化对性能有一些影响。因此建议确保手眼标定过程中的温度在一定程度上保持稳定。这可以通过 预热 程序实现。最好选择与实际部署类似的工作环境进行标定。如需进一步降低与温度相关的性能因素的影响，请启用 热稳定功能。

选择正确的方法

Some hand-eye calibration methods compute camera intrinsic parameters along with extrinsic parameters and the relative pose between the camera and the robot frame. We do not recommend these approaches because they treat a Zivid camera as an uncalibrated 2D camera, rather than a well-calibrated 3D camera.

Each Zivid camera unit goes through an extensive calibration process, which includes determining the intrinsic parameters of its 2D color camera. Our calibration uses a complex camera model with more intrinsic parameters than some well-known pinhole camera models, e.g., OpenCV camera model. Since Zivid camera model is proprietary, our internal camera intrinsics are not available in the SDK. However, Zivid SDK does offer approximations of OpenCV and Halcon models (see Camera Intrinsics) from our camera model. Because information is lost in approximation, using hand-eye calibration methods that utilize OpenCV or Halcon intrinsics is not the best approach either.

Since Zivid cameras provide 3D data, it is possible to calculate camera extrinsics from the point cloud. Zivid Hand-Eye Calibration method utilizes this benefit and that is why it is the recommended approach and the one that works best with our cameras. There are alternative, non-Zivid methods, that utilize the possibility to calculate camera extrinsics from the point cloud. These methods rely purely on point cloud data and an example is hand-eye calibration based on CAD matching.

准确度和重新标定

The picking accuracy of a vision-guided robotic system depends on the combined accuracy of the camera, hand-eye calibration, machine vision software, and robot’s positioning. Robots are in general highly repeatable but not accurate. Temperature, joint friction, payload, and manufacturing tolerances are some of the factors that cause the robot to deviate from its preprogrammed positioning. However, robot pose accuracy can be improved by calibrating the robot itself, which is highly recommended for complex systems with multiple factors that affect the picking accuracy. If the robot loses the calibration, the picking accuracy will deteriorate. Repeating the calibration (robot and/or hand-eye) can compensate for such deteriorated performance. It is also necessary to repeat the hand-eye calibration after dismounting the camera from a fixed structure or a robot and mounting it back on.

阅读更多关于 手眼标定的残差.