factor confidence/covariance in gtsam NonlinearFactorGraph

[Aniket Kumar Roy | Robotics Engineer]

Hi

I am using gtsam for slam using its isam2 implementation. the variables for the Non-linear factor graph are the poses with respect to the world represented as a gtsam::Similarity3 structure, the in between factors are wheel odometry readings and visual odometry calculated through feature detection and matching from each camera frame. the wheel odometry readings are relative movements with respect to previous pose represented as a gtsam::Similarity3 structure, same case for visual odometry factor representing relative change with respect to previous frame.

now I want a way to measure the confidence gtsam have for a key's pose or the confidence it have for a particular factor after isam update. is there any utility function in gtsam that can give me that?

Thanks in advance

[Brice Rebsamen]

hello,
yes, you have to check the marginals. I can't really check the doc at
the moment, but search for Marginal and you'll find it
Brice

> --
> You received this message because you are subscribed to the Google Groups "gtsam users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to gtsam-users...@googlegroups.com.
> To view this discussion visit https://groups.google.com/d/msgid/gtsam-users/65e25e63-cc08-459c-90b3-eaab5f3280e0n%40googlegroups.com.