TrackedBodyTarget.cpp File Reference

Implementation. More...

#include "TrackedBodyTarget.h"
#include "AssignMeasurementsToLeds.h"
#include "BodyTargetInterface.h"
#include "HDKLedIdentifier.h"
#include "LED.h"
#include "PoseEstimatorTypes.h"
#include "PoseEstimator_RANSAC.h"
#include "PoseEstimator_RANSACKalman.h"
#include "PoseEstimator_SCAATKalman.h"
#include "TrackedBody.h"
#include "cvToEigen.h"
#include <osvr/Util/CSV.h>
#include <osvr/Util/CSVCellGroup.h>
#include <boost/assert.hpp>
#include <util/Stride.h>
#include <algorithm>
#include <fstream>
#include <iostream>

Classes
class	osvr::vbtracker::TargetHealthEvaluator
struct	osvr::vbtracker::TrackedBodyTarget::Impl

Namespaces
	osvr
	The main namespace for all C++ elements of the framework, internal and external.

Enumerations
enum	TargetTrackingState {   RANSAC, RANSACKalman, EnteringKalman, Kalman,   RANSACWhenBlobDetected, RANSACKalmanWhenBlobDetected }
enum	TargetHealthState { OK, StopTrackingErrorBoundsExceeded, StopTrackingLostSight, HardResetNonFiniteState }

Functions
double	osvr::vbtracker::getLimitOnMaxPositionalErrorVariance (double distance)
double	osvr::vbtracker::getMaxPositionalErrorVariance (BodyState const &bodyState)
bool	osvr::vbtracker::isStateSCAAT (TargetTrackingState trackingState)
	Predicate to check simply if the state corresponds to running the SCAAT Kalman filter mode, centralized and named. More...
BeaconStateVec	osvr::vbtracker::createBeaconStateVec (ConfigParams const &params, TargetSetupData const &setupData, Eigen::Vector3d &beaconOffset)
std::ptrdiff_t	osvr::vbtracker::getNumUsedLeds (LedPtrList const &usableLeds)

Detailed Description

Implementation.

Date

2016

Author

Sensics, Inc. http://sensics.com/osvr

Function Documentation

createBeaconStateVec()

BeaconStateVec osvr::vbtracker::createBeaconStateVec ( ConfigParams const &  params, TargetSetupData const &  setupData, Eigen::Vector3d &  beaconOffset  )

inline

Compute or retrieve the beacon offset.

Create the vector we'll return, and then the beacon state objects.

getLimitOnMaxPositionalErrorVariance()

double osvr::vbtracker::getLimitOnMaxPositionalErrorVariance ( double  distance )

inline

An exponential function was fit to recorded data of max positional error variances and distances during normal tracking operation, using libreoffice calc. R^2 = 0.95, and on a plot with a log-scale for y, the data looked fairly linear.

log-plot slope B is used directly from that regression. coefficient A hand-picked based on original fit, to move this function above the normal data with a good margin of error.

isStateSCAAT()

bool osvr::vbtracker::isStateSCAAT ( TargetTrackingState  trackingState )

inline

Predicate to check simply if the state corresponds to running the SCAAT Kalman filter mode, centralized and named.