For the past few weeks I have been looking at the EKF-SLAM inconsistency problem. I've compiled what I've found so far into a short literature review linked here . This includes some references for proposed modifications to the EKF-SLAM algorithm. Based on what I've looked at so far, the ways to remedy the EKF-SLAM inconsistency problem with localization include making smaller maps and joining these together or changing the mathematical formulae for the EKF algorithm to change the observability of the algorithm to match the problem subspace. To more fully understand what is happening here, more research on observability should be done.

Next Post Previous Post