I am trying to optimize a pose-graph generated through visual odometry with loop closure constraints.

I wanted to know if there are any particular situations where optimizing the graph only once is beneficial as opposed to optimizing it everytime upon finding a loop closure. I am not sure what happens to the original constraints once an optimization step is performed (I am using g2o).

Whenever I try to optimize continuously, my pose graph actually deviates farther from the ground truth. Also there is no difference between the initial and the final loss.

I wanted to know how to incorporate new constraints after an optimization step is complete?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.