multiview registration without odometry edge.

[songhc8]

Checklist

• [X] I have searched for similar issues.
• [X] For Python issues, I have tested with the latest development wheel.
• [X] I have checked the release documentation and the latest documentation (for main branch).

My Question

To whom it may concern,

My camera is mounted on the robot, and hand-eye calibration can obtain the extrinsic parameters between the robot and the camera. Suppose I have a set of scans at different positions, scan1-scanN. For each scan, I can directly read out the camera's pose P_i. the transformation between pose i and j can be got H_i2j=P_j.inverse()*P_i. it also can be calculated from scans_i and scan_j by fpfh algorithm or other methods. i.e. H_est_i2j=fpfh(scan_i,scan_j)

Due to the errors in hand-eye calibration, the scans are not perfectly aligned when transform them into robot's base frame using H_i2j.

I would like to build pose graph to optimize the pose. When constructing the pose graph, should the edges be set to H_i2j, and uncertain set to false? Or should the edges be set to H_est_i2j, and uncertain set to ture.

Thanks for you reply.