About Continuous-Time Trajectory (Under development)

[Lurvelly]

Hi, koide,
I found you write ContinuousTrajectory Cubic B-Spline-based continuous trajectory representation for offline batch optimization in your README, but I can't found this paper. Is this paper not published yet?

[koide3]

It is basically a re-implementation of [Sommer+, CVP2020], and there is no publication dedicated to it so far. Meanwhile, I'm writing a technical report that summarizes the features implemented in gtsam_points that may involve a short description of the implemented approach.

[Lurvelly]

That's great. I'm looking forward to your technical report. By the way, have you considered the Gaussian Process for the Continuous-Time Trajectory? It seems better than B-Spline representation ref to GPTR: Gaussian Process Trajectory Representation
for Continuous-Time Motion Estimation.

[koide3]

I'm not very familiar with the GP-based CT trajectory, but I feel it may be a bit costly (like other GP-based methods). I think the usual CT representation would suffice for most applications.