Description of the task
We set up kalman filter variables like model matrices, covariance matrices in different ways:
-
In ball filter, we use a comma initialzer to set up our Eigen matrices:
-
-
-
In robot localizer, we set up using coefficients:
-
m(XXCOVARIANCE, XYCOVARIANCE) = 3
The first method is not really safe on initialize. If we accidentally input an extra value we won't be able to catch it.
The second method creates a giant block of code in the middle of the robot localizer, costing our code quality and readibility.
Explore whether there are alternative ways to configure these variables. Maybe we configure them in a separate file and load them in?
Acceptance criteria
Blocked By
Description of the task
We set up kalman filter variables like model matrices, covariance matrices in different ways:
In ball filter, we use a comma initialzer to set up our Eigen matrices:
Eigen::Matrix2d m;m << 3 , 3, 3, 5;In robot localizer, we set up using coefficients:
m(XXCOVARIANCE, XYCOVARIANCE) = 3The first method is not really safe on initialize. If we accidentally input an extra value we won't be able to catch it.
The second method creates a giant block of code in the middle of the robot localizer, costing our code quality and readibility.
Explore whether there are alternative ways to configure these variables. Maybe we configure them in a separate file and load them in?
Acceptance criteria
Blocked By