constrained_solvepnp Namespace Reference

Classes
struct	CameraCalibration
	Pinhole camera coefficients. More...

Typedefs
using	casadi_real = double
using	RobotStateMat = Eigen::Matrix<casadi_real, 3, 1>

Functions
wpi::expected< RobotStateMat, sleipnir::SolverExitCondition >	do_optimization (bool heading_free, int nTags, CameraCalibration cameraCal, Eigen::Matrix< casadi_real, 4, 4, Eigen::ColMajor > robot2camera, RobotStateMat x_guess, Eigen::Matrix< casadi_real, 4, Eigen::Dynamic, Eigen::ColMajor > field2points, Eigen::Matrix< casadi_real, 2, Eigen::Dynamic, Eigen::ColMajor > point_observations, double gyroθ, double gyroErrorScaleFac)
	Optimize x, where x is [x, y, theta]^T.

Typedef Documentation

casadi_real

using constrained_solvepnp::casadi_real = double

RobotStateMat

using constrained_solvepnp::RobotStateMat = Eigen::Matrix<casadi_real, 3, 1>

Function Documentation

do_optimization()

wpi::expected< RobotStateMat, sleipnir::SolverExitCondition > constrained_solvepnp::do_optimization	(	bool	heading_free,
		int	nTags,
		CameraCalibration	cameraCal,
		Eigen::Matrix< casadi_real, 4, 4, Eigen::ColMajor >	robot2camera,
		RobotStateMat	x_guess,
		Eigen::Matrix< casadi_real, 4, Eigen::Dynamic, Eigen::ColMajor >	field2points,
		Eigen::Matrix< casadi_real, 2, Eigen::Dynamic, Eigen::ColMajor >	point_observations,
		double	gyroθ,
		double	gyroErrorScaleFac )

Optimize x, where x is [x, y, theta]^T.

Note points must be undistorted prior to this. The number of columns in field2points and point_observations just be exactly 4x nTags.