Our new paper, Projection Filtering With Observed State Increments With Applications in Continuous-Time Circular Filtering by Anna Kutschireiter and Luke Rast, just appeared in IEEE Transactions on Signal Processing.

In that paper we derive a new projection filter that is able to handle state increment observations in addition to the more traditional direct state observations. Applied to angular path integration we derive a set of closed-form equations that we call the circular Kalman filter.

For more details, check Anna’s Twitter post.