MotionModel.hpp File Reference

#include <iostream>
#include <vector>
#include <boost/shared_ptr.hpp>
#include <base-logging/Logging.hpp>
#include <Eigen/Geometry>
#include <Eigen/Core>
#include <Eigen/Dense>
#include <Eigen/Cholesky>

Go to the source code of this file.

Classes
class	threed_odometry::MotionModel< _Scalar >

Namespaces
	threed_odometry

Detailed Description

This class provided a Motion Model solver for any mobile robot (wheel or leg) with articulated joints.

The solver is based on Weighted Least-Squares as minimizing the error to estimate the resulting motion from a robot Jacobian. Robot Jacobian is understood as the sparse Jacobian matrix containing one Jacobian per each Chain of the Robot. Therefore, it is a minimization problem of a non-linear system which has been previously linearized around the working point (robot current joint position values). This is represented in the current Jacobian matrix of the robot. This class requires a robot Jacobian and computes the statistical motion model (estimated velocities/navigation quantities).

Further Details at: P.Muir et. al Kinematic Modeling of Wheeled Mobile Robots M. Tarokh et. al Kinematics Modeling and Analyses of Articulated Rovers J. Hidalgo et. al Kinematics Modeling of a Hybrid Wheeled-Leg Planetary Rover J. Hidalgo, Navigation and Slip Kinematics for High Performance Motion Models Hidalgo-Carrio, Javier and Babu, Ajish and Kirchner, Frank, Static forces weighted Jacobian motion models for improved Odometry

Author

Javier Hidalgo Carrio | DFKI RIC Bremen | javie.nosp@m.r.hi.nosp@m.dalgo.nosp@m._car.nosp@m.rio@d.nosp@m.fki..nosp@m.de

Date

December 2014.

Version

1.0.