fanfufei / lse

Legged State Estimation Library

LSE

This is the Legged State Estimator (LSE) library. It contains an Observability Constrained Extended Kalman Filter and a routine for time delay calibration between different sensor modalities. The Manager class represents the interface. It handles the different measurements.

INSTALLATION:

• go to top folder (the folder this file is located in)
• execute cmake: "cmake ."
• either simply compile the library: "make", or call the installation: "sudo make install" (this will copy the library and header to your system folder and create a corresponding FindLSE.cmake file)

DEPENDENCIES:

• Standard library
• Eigen3
• Python (2.7)
• Ceres (1.5.0)
• The required tinyxml library is provided together with this installation

TODO:

• Clean up with old filters
• Clean up quaternion/rotation stuff
• Clean up time delay stuff