EKF SLAM (prediction of new landmarks)

Question

Prediction of new landmarks are commonly expressed as:

        Xm = Xr + r*cos(phi + theta_r),
        Ym = Yr + r*sin(phi + theta_r)

However this is only true for point landmarks. What if I am extracting line feature?

Answer 1

This is true because landmarks are often encapsulated by just their coordinates. It's a toy example. In real life, you'd have to use all the parameters necessary to encapsulate the information about the line feature. For example. Lines are often represented by a point and orientation.

• here's an example