Compare commits

...

13 Commits

Author SHA1 Message Date
2dac361ba2 fixed stop - go functionality via bagcontrol 2019-06-28 09:40:59 +02:00
0712a98c7f Merge branch 'photometry-jakobi-htest' of https://gitlab.com/Hippocampus/msckf into photometry-jakobi-htest 2019-06-25 12:08:41 +02:00
6b8dce9876 removed constant logging 2019-06-19 09:29:22 +02:00
49374a4323 added direct rho estimation 2019-06-14 12:37:06 +02:00
3e480560e8 Update msckf_vio.cpp 2019-06-12 09:25:38 +00:00
b3df525060 made residual change more direct. starts diverging but then sometimes appriximates correctly after some camera movement 2019-06-12 09:26:34 +02:00
a8d4580812 removed some groundtruth measurement processings 2019-06-11 17:03:05 +02:00
a0577dfb9d added some feature calculations to this branch 2019-06-11 16:45:17 +02:00
8cfbe06945 minor calcualtion changes - works at 7-8 fps bagfiles speed 2019-05-28 14:12:22 +02:00
cab56d9494 tested bug: filter converges longer without distortion effects; reason probably general formulation inconsistencies 2019-05-22 10:57:31 +02:00
5e9149eacc constructed jacobian based on residual: r = z(measurement pix. pos) - h(rho, x_l, x_a) - based on the projection of the feature point into space based on the anchor frame - followed by projection onto H_rho nullspace; works well enough 2019-05-21 14:50:53 +02:00
e4dbe2f060 removed error in photom. res. calculation 2019-05-21 14:23:49 +02:00
6b208dbc44 added changed formulation, no positive result 2019-05-16 15:53:07 +02:00
7 changed files with 601 additions and 247 deletions

View File

@ -15,7 +15,7 @@
#include <Eigen/Dense> #include <Eigen/Dense>
#include <Eigen/Geometry> #include <Eigen/Geometry>
#include <Eigen/StdVector> #include <Eigen/StdVector>
#include <math.h>
#include <visualization_msgs/Marker.h> #include <visualization_msgs/Marker.h>
#include <visualization_msgs/MarkerArray.h> #include <visualization_msgs/MarkerArray.h>
#include <geometry_msgs/Point.h> #include <geometry_msgs/Point.h>
@ -70,6 +70,11 @@ struct Feature {
position(Eigen::Vector3d::Zero()), position(Eigen::Vector3d::Zero()),
is_initialized(false), is_anchored(false) {} is_initialized(false), is_anchored(false) {}
void Rhocost(const Eigen::Isometry3d& T_c0_ci,
const double x, const Eigen::Vector2d& z1, const Eigen::Vector2d& z2,
double& e) const;
/* /*
* @brief cost Compute the cost of the camera observations * @brief cost Compute the cost of the camera observations
* @param T_c0_c1 A rigid body transformation takes * @param T_c0_c1 A rigid body transformation takes
@ -82,6 +87,13 @@ struct Feature {
const Eigen::Vector3d& x, const Eigen::Vector2d& z, const Eigen::Vector3d& x, const Eigen::Vector2d& z,
double& e) const; double& e) const;
bool initializeRho(const CamStateServer& cam_states);
inline void RhoJacobian(const Eigen::Isometry3d& T_c0_ci,
const double x, const Eigen::Vector2d& z1, const Eigen::Vector2d& z2,
Eigen::Matrix<double, 2, 1>& J, Eigen::Vector2d& r,
double& w) const;
/* /*
* @brief jacobian Compute the Jacobian of the camera observation * @brief jacobian Compute the Jacobian of the camera observation
* @param T_c0_c1 A rigid body transformation takes * @param T_c0_c1 A rigid body transformation takes
@ -97,6 +109,10 @@ struct Feature {
Eigen::Matrix<double, 2, 3>& J, Eigen::Vector2d& r, Eigen::Matrix<double, 2, 3>& J, Eigen::Vector2d& r,
double& w) const; double& w) const;
inline double generateInitialDepth(
const Eigen::Isometry3d& T_c1_c2, const Eigen::Vector2d& z1,
const Eigen::Vector2d& z2) const;
/* /*
* @brief generateInitialGuess Compute the initial guess of * @brief generateInitialGuess Compute the initial guess of
* the feature's 3d position using only two views. * the feature's 3d position using only two views.
@ -148,6 +164,14 @@ struct Feature {
inline bool initializePosition( inline bool initializePosition(
const CamStateServer& cam_states); const CamStateServer& cam_states);
cv::Point2f pixelDistanceAt(
const CAMState& cam_state,
const StateIDType& cam_state_id,
const CameraCalibration& cam,
Eigen::Vector3d& in_p) const;
/* /*
* @brief project PositionToCamera Takes a 3d position in a world frame * @brief project PositionToCamera Takes a 3d position in a world frame
* and projects it into the passed camera frame using pinhole projection * and projects it into the passed camera frame using pinhole projection
@ -160,6 +184,11 @@ struct Feature {
const CameraCalibration& cam, const CameraCalibration& cam,
Eigen::Vector3d& in_p) const; Eigen::Vector3d& in_p) const;
double Kernel(
const cv::Point2f pose,
const cv::Mat frame,
std::string type) const;
/* /*
* @brief IrradianceAnchorPatch_Camera returns irradiance values * @brief IrradianceAnchorPatch_Camera returns irradiance values
* of the Anchor Patch position in a camera frame * of the Anchor Patch position in a camera frame
@ -181,13 +210,15 @@ bool MarkerGeneration(
const CAMState& cam_state, const CAMState& cam_state,
const StateIDType& cam_state_id, const StateIDType& cam_state_id,
CameraCalibration& cam0, CameraCalibration& cam0,
const std::vector<double> photo_r, const Eigen::VectorXd& photo_r,
std::stringstream& ss) const; std::stringstream& ss,
cv::Point2f gradientVector,
cv::Point2f residualVector) const;
/* /*
* @brief projectPixelToPosition uses the calcualted pixels * @brief AnchorPixelToPosition uses the calcualted pixels
* of the anchor patch to generate 3D positions of all of em * of the anchor patch to generate 3D positions of all of em
*/ */
inline Eigen::Vector3d projectPixelToPosition(cv::Point2f in_p, inline Eigen::Vector3d AnchorPixelToPosition(cv::Point2f in_p,
const CameraCalibration& cam); const CameraCalibration& cam);
/* /*
@ -248,6 +279,26 @@ typedef std::map<FeatureIDType, Feature, std::less<int>,
Eigen::aligned_allocator< Eigen::aligned_allocator<
std::pair<const FeatureIDType, Feature> > > MapServer; std::pair<const FeatureIDType, Feature> > > MapServer;
void Feature::Rhocost(const Eigen::Isometry3d& T_c0_ci,
const double x, const Eigen::Vector2d& z1, const Eigen::Vector2d& z2,
double& e) const
{
// Compute hi1, hi2, and hi3 as Equation (37).
const double& rho = x;
Eigen::Vector3d h = T_c0_ci.linear()*
Eigen::Vector3d(z1(0), z1(1), 1.0) + rho*T_c0_ci.translation();
double& h1 = h(0);
double& h2 = h(1);
double& h3 = h(2);
// Predict the feature observation in ci frame.
Eigen::Vector2d z_hat(h1/h3, h2/h3);
// Compute the residual.
e = (z_hat-z2).squaredNorm();
return;
}
void Feature::cost(const Eigen::Isometry3d& T_c0_ci, void Feature::cost(const Eigen::Isometry3d& T_c0_ci,
const Eigen::Vector3d& x, const Eigen::Vector2d& z, const Eigen::Vector3d& x, const Eigen::Vector2d& z,
@ -260,9 +311,9 @@ void Feature::cost(const Eigen::Isometry3d& T_c0_ci,
Eigen::Vector3d h = T_c0_ci.linear()* Eigen::Vector3d h = T_c0_ci.linear()*
Eigen::Vector3d(alpha, beta, 1.0) + rho*T_c0_ci.translation(); Eigen::Vector3d(alpha, beta, 1.0) + rho*T_c0_ci.translation();
double& h1 = h(0); double h1 = h(0);
double& h2 = h(1); double h2 = h(1);
double& h3 = h(2); double h3 = h(2);
// Predict the feature observation in ci frame. // Predict the feature observation in ci frame.
Eigen::Vector2d z_hat(h1/h3, h2/h3); Eigen::Vector2d z_hat(h1/h3, h2/h3);
@ -272,6 +323,42 @@ void Feature::cost(const Eigen::Isometry3d& T_c0_ci,
return; return;
} }
void Feature::RhoJacobian(const Eigen::Isometry3d& T_c0_ci,
const double x, const Eigen::Vector2d& z1, const Eigen::Vector2d& z2,
Eigen::Matrix<double, 2, 1>& J, Eigen::Vector2d& r,
double& w) const
{
const double& rho = x;
Eigen::Vector3d h = T_c0_ci.linear()*
Eigen::Vector3d(z1(0), z2(1), 1.0) + rho*T_c0_ci.translation();
double& h1 = h(0);
double& h2 = h(1);
double& h3 = h(2);
// Compute the Jacobian.
Eigen::Matrix3d W;
W.leftCols<2>() = T_c0_ci.linear().leftCols<2>();
W.rightCols<1>() = T_c0_ci.translation();
J(0,0) = -h1/(h3*h3);
J(1,0) = -h2/(h3*h3);
// Compute the residual.
Eigen::Vector2d z_hat(h1/h3, h2/h3);
r = z_hat - z2;
// Compute the weight based on the residual.
double e = r.norm();
if (e <= optimization_config.huber_epsilon)
w = 1.0;
else
w = optimization_config.huber_epsilon / (2*e);
return;
}
void Feature::jacobian(const Eigen::Isometry3d& T_c0_ci, void Feature::jacobian(const Eigen::Isometry3d& T_c0_ci,
const Eigen::Vector3d& x, const Eigen::Vector2d& z, const Eigen::Vector3d& x, const Eigen::Vector2d& z,
Eigen::Matrix<double, 2, 3>& J, Eigen::Vector2d& r, Eigen::Matrix<double, 2, 3>& J, Eigen::Vector2d& r,
@ -311,9 +398,9 @@ void Feature::jacobian(const Eigen::Isometry3d& T_c0_ci,
return; return;
} }
void Feature::generateInitialGuess( double Feature::generateInitialDepth(
const Eigen::Isometry3d& T_c1_c2, const Eigen::Vector2d& z1, const Eigen::Isometry3d& T_c1_c2, const Eigen::Vector2d& z1,
const Eigen::Vector2d& z2, Eigen::Vector3d& p) const const Eigen::Vector2d& z2) const
{ {
// Construct a least square problem to solve the depth. // Construct a least square problem to solve the depth.
Eigen::Vector3d m = T_c1_c2.linear() * Eigen::Vector3d(z1(0), z1(1), 1.0); Eigen::Vector3d m = T_c1_c2.linear() * Eigen::Vector3d(z1(0), z1(1), 1.0);
@ -328,6 +415,15 @@ void Feature::generateInitialGuess(
// Solve for the depth. // Solve for the depth.
double depth = (A.transpose() * A).inverse() * A.transpose() * b; double depth = (A.transpose() * A).inverse() * A.transpose() * b;
return depth;
}
void Feature::generateInitialGuess(
const Eigen::Isometry3d& T_c1_c2, const Eigen::Vector2d& z1,
const Eigen::Vector2d& z2, Eigen::Vector3d& p) const
{
double depth = generateInitialDepth(T_c1_c2, z1, z2);
p(0) = z1(0) * depth; p(0) = z1(0) * depth;
p(1) = z1(1) * depth; p(1) = z1(1) * depth;
p(2) = depth; p(2) = depth;
@ -377,6 +473,26 @@ bool Feature::checkMotion(const CamStateServer& cam_states) const
else return false; else return false;
} }
double Feature::Kernel(
const cv::Point2f pose,
const cv::Mat frame,
std::string type) const
{
Eigen::Matrix<double, 3, 3> kernel = Eigen::Matrix<double, 3, 3>::Zero();
if(type == "Sobel_x")
kernel << -1., 0., 1.,-2., 0., 2. , -1., 0., 1.;
else if(type == "Sobel_y")
kernel << -1., -2., -1., 0., 0., 0., 1., 2., 1.;
double delta = 0;
int offs = (int)(kernel.rows()-1)/2;
for(int i = 0; i < kernel.rows(); i++)
for(int j = 0; j < kernel.cols(); j++)
delta += ((float)frame.at<uint8_t>(pose.y+j-offs , pose.x+i-offs))/255 * (float)kernel(j,i);
return delta;
}
bool Feature::estimate_FrameIrradiance( bool Feature::estimate_FrameIrradiance(
const CAMState& cam_state, const CAMState& cam_state,
const StateIDType& cam_state_id, const StateIDType& cam_state_id,
@ -532,8 +648,10 @@ bool Feature::VisualizePatch(
const CAMState& cam_state, const CAMState& cam_state,
const StateIDType& cam_state_id, const StateIDType& cam_state_id,
CameraCalibration& cam0, CameraCalibration& cam0,
const std::vector<double> photo_r, const Eigen::VectorXd& photo_r,
std::stringstream& ss) const std::stringstream& ss,
cv::Point2f gradientVector,
cv::Point2f residualVector) const
{ {
double rescale = 1; double rescale = 1;
@ -573,44 +691,106 @@ bool Feature::VisualizePatch(
cv::hconcat(cam0.featureVisu, dottedFrame, cam0.featureVisu); cv::hconcat(cam0.featureVisu, dottedFrame, cam0.featureVisu);
// irradiance grid anchor
// patches visualization
int N = sqrt(anchorPatch_3d.size()); int N = sqrt(anchorPatch_3d.size());
int scale = 20; int scale = 30;
cv::Mat irradianceFrame(anchorImage.size(), CV_8UC3, cv::Scalar(255, 240, 255)); cv::Mat irradianceFrame(anchorImage.size(), CV_8UC3, cv::Scalar(255, 240, 255));
cv::resize(irradianceFrame, irradianceFrame, cv::Size(), rescale, rescale); cv::resize(irradianceFrame, irradianceFrame, cv::Size(), rescale, rescale);
// irradiance grid anchor
std::stringstream namer;
namer << "anchor";
cv::putText(irradianceFrame, namer.str() , cvPoint(30, 25),
cv::FONT_HERSHEY_COMPLEX_SMALL, 0.8, cvScalar(0,0,0), 1, CV_AA);
for(int i = 0; i<N; i++) for(int i = 0; i<N; i++)
for(int j = 0; j<N; j++) for(int j = 0; j<N; j++)
cv::rectangle(irradianceFrame, cv::rectangle(irradianceFrame,
cv::Point(10+scale*(i+1), 10+scale*j), cv::Point(30+scale*(i+1), 30+scale*j),
cv::Point(10+scale*i, 10+scale*(j+1)), cv::Point(30+scale*i, 30+scale*(j+1)),
cv::Scalar(anchorPatch[i*N+j]*255, anchorPatch[i*N+j]*255, anchorPatch[i*N+j]*255), cv::Scalar(anchorPatch[i*N+j]*255, anchorPatch[i*N+j]*255, anchorPatch[i*N+j]*255),
CV_FILLED); CV_FILLED);
// irradiance grid projection // irradiance grid projection
namer.str(std::string());
namer << "projection";
cv::putText(irradianceFrame, namer.str() , cvPoint(30, 45+scale*N),
cv::FONT_HERSHEY_COMPLEX_SMALL, 0.8, cvScalar(0,0,0), 1, CV_AA);
for(int i = 0; i<N; i++) for(int i = 0; i<N; i++)
for(int j = 0; j<N ; j++) for(int j = 0; j<N ; j++)
cv::rectangle(irradianceFrame, cv::rectangle(irradianceFrame,
cv::Point(10+scale*(i+1), 20+scale*(N+j)), cv::Point(30+scale*(i+1), 50+scale*(N+j)),
cv::Point(10+scale*(i), 20+scale*(N+j+1)), cv::Point(30+scale*(i), 50+scale*(N+j+1)),
cv::Scalar(projectionPatch[i*N+j]*255, projectionPatch[i*N+j]*255, projectionPatch[i*N+j]*255), cv::Scalar(projectionPatch[i*N+j]*255, projectionPatch[i*N+j]*255, projectionPatch[i*N+j]*255),
CV_FILLED); CV_FILLED);
// true irradiance at feature
// get current observation
namer.str(std::string());
namer << "feature";
cv::putText(irradianceFrame, namer.str() , cvPoint(30, 65+scale*2*N),
cv::FONT_HERSHEY_COMPLEX_SMALL, 0.8, cvScalar(0,0,0), 1, CV_AA);
cv::Point2f p_f(observations.find(cam_state_id)->second(0),observations.find(cam_state_id)->second(1));
// move to real pixels
p_f = image_handler::distortPoint(p_f, cam0.intrinsics, cam0.distortion_model, cam0.distortion_coeffs);
for(int i = 0; i<N; i++)
{
for(int j = 0; j<N ; j++)
{
float irr = PixelIrradiance(cv::Point2f(p_f.x + (i-(N-1)/2), p_f.y + (j-(N-1)/2)), current_image);
cv::rectangle(irradianceFrame,
cv::Point(30+scale*(i+1), 70+scale*(2*N+j)),
cv::Point(30+scale*(i), 70+scale*(2*N+j+1)),
cv::Scalar(irr*255, irr*255, irr*255),
CV_FILLED);
}
}
// residual grid projection, positive - red, negative - blue colored // residual grid projection, positive - red, negative - blue colored
namer.str(std::string());
namer << "residual";
cv::putText(irradianceFrame, namer.str() , cvPoint(30+scale*N, scale*N/2-5),
cv::FONT_HERSHEY_COMPLEX_SMALL, 0.8, cvScalar(0,0,0), 1, CV_AA);
for(int i = 0; i<N; i++) for(int i = 0; i<N; i++)
for(int j = 0; j<N; j++) for(int j = 0; j<N; j++)
if(photo_r[i*N+j]>0) if(photo_r(i*N+j)>0)
cv::rectangle(irradianceFrame, cv::rectangle(irradianceFrame,
cv::Point(20+scale*(N+i+1), 15+scale*(N/2+j)), cv::Point(40+scale*(N+i+1), 15+scale*(N/2+j)),
cv::Point(20+scale*(N+i), 15+scale*(N/2+j+1)), cv::Point(40+scale*(N+i), 15+scale*(N/2+j+1)),
cv::Scalar(255 - photo_r[i*N+j]*255, 255 - photo_r[i*N+j]*255, 255), cv::Scalar(255 - photo_r(i*N+j)*255, 255 - photo_r(i*N+j)*255, 255),
CV_FILLED); CV_FILLED);
else else
cv::rectangle(irradianceFrame, cv::rectangle(irradianceFrame,
cv::Point(20+scale*(N+i+1), 15+scale*(N/2+j)), cv::Point(40+scale*(N+i+1), 15+scale*(N/2+j)),
cv::Point(20+scale*(N+i), 15+scale*(N/2+j+1)), cv::Point(40+scale*(N+i), 15+scale*(N/2+j+1)),
cv::Scalar(255, 255 + photo_r[i*N+j]*255, 255 + photo_r[i*N+j]*255), cv::Scalar(255, 255 + photo_r(i*N+j)*255, 255 + photo_r(i*N+j)*255),
CV_FILLED); CV_FILLED);
// gradient arrow
/*
cv::arrowedLine(irradianceFrame,
cv::Point(30+scale*(N/2 +0.5), 50+scale*(N+(N/2)+0.5)),
cv::Point(30+scale*(N/2+0.5)+scale*gradientVector.x, 50+scale*(N+(N/2)+0.5)+scale*gradientVector.y),
cv::Scalar(100, 0, 255),
1);
*/
// residual gradient direction
cv::arrowedLine(irradianceFrame,
cv::Point(40+scale*(N+N/2+0.5), 15+scale*((N-0.5))),
cv::Point(40+scale*(N+N/2+0.5)+scale*residualVector.x, 15+scale*(N-0.5)+scale*residualVector.y),
cv::Scalar(0, 255, 175),
3);
cv::hconcat(cam0.featureVisu, irradianceFrame, cam0.featureVisu); cv::hconcat(cam0.featureVisu, irradianceFrame, cam0.featureVisu);
/* /*
@ -671,6 +851,37 @@ float Feature::PixelIrradiance(cv::Point2f pose, cv::Mat image) const
return ((float)image.at<uint8_t>(pose.y, pose.x))/255; return ((float)image.at<uint8_t>(pose.y, pose.x))/255;
} }
cv::Point2f Feature::pixelDistanceAt(
const CAMState& cam_state,
const StateIDType& cam_state_id,
const CameraCalibration& cam,
Eigen::Vector3d& in_p) const
{
cv::Point2f cam_p = projectPositionToCamera(cam_state, cam_state_id, cam, in_p);
// create vector of patch in pixel plane
std::vector<cv::Point2f> surroundingPoints;
surroundingPoints.push_back(cv::Point2f(cam_p.x+1, cam_p.y));
surroundingPoints.push_back(cv::Point2f(cam_p.x-1, cam_p.y));
surroundingPoints.push_back(cv::Point2f(cam_p.x, cam_p.y+1));
surroundingPoints.push_back(cv::Point2f(cam_p.x, cam_p.y-1));
std::vector<cv::Point2f> pure;
image_handler::undistortPoints(surroundingPoints,
cam.intrinsics,
cam.distortion_model,
cam.distortion_coeffs,
pure);
// returns the absolute pixel distance at pixels one metres away
cv::Point2f distance(fabs(pure[0].x - pure[1].x), fabs(pure[2].y - pure[3].y));
return distance;
}
cv::Point2f Feature::projectPositionToCamera( cv::Point2f Feature::projectPositionToCamera(
const CAMState& cam_state, const CAMState& cam_state,
const StateIDType& cam_state_id, const StateIDType& cam_state_id,
@ -703,7 +914,7 @@ cv::Point2f Feature::projectPositionToCamera(
return my_p; return my_p;
} }
Eigen::Vector3d Feature::projectPixelToPosition(cv::Point2f in_p, const CameraCalibration& cam) Eigen::Vector3d Feature::AnchorPixelToPosition(cv::Point2f in_p, const CameraCalibration& cam)
{ {
// use undistorted position of point of interest // use undistorted position of point of interest
// project it back into 3D space using pinhole model // project it back into 3D space using pinhole model
@ -742,27 +953,19 @@ bool Feature::initializeAnchor(const CameraCalibration& cam, int N)
cv::Point2f und_pix_p = image_handler::distortPoint(cv::Point2f(u, v), cv::Point2f und_pix_p = image_handler::distortPoint(cv::Point2f(u, v),
cam.intrinsics, cam.intrinsics,
cam.distortion_model, cam.distortion_model,
0); cam.distortion_coeffs);
// create vector of patch in pixel plane // create vector of patch in pixel plane
std::vector<cv::Point2f>und_pix_v;
for(double u_run = -n; u_run <= n; u_run++) for(double u_run = -n; u_run <= n; u_run++)
for(double v_run = -n; v_run <= n; v_run++) for(double v_run = -n; v_run <= n; v_run++)
und_pix_v.push_back(cv::Point2f(und_pix_p.x+u_run, und_pix_p.y+v_run)); anchorPatch_real.push_back(cv::Point2f(und_pix_p.x+u_run, und_pix_p.y+v_run));
//create undistorted pure points //create undistorted pure points
std::vector<cv::Point2f> und_v; image_handler::undistortPoints(anchorPatch_real,
image_handler::undistortPoints(und_pix_v,
cam.intrinsics, cam.intrinsics,
cam.distortion_model, cam.distortion_model,
0, cam.distortion_coeffs,
und_v); anchorPatch_ideal);
//create distorted pixel points
anchorPatch_real = image_handler::distortPoints(und_v,
cam.intrinsics,
cam.distortion_model,
cam.distortion_coeffs);
// save anchor position for later visualisaztion // save anchor position for later visualisaztion
@ -770,26 +973,170 @@ bool Feature::initializeAnchor(const CameraCalibration& cam, int N)
// save true pixel Patch position // save true pixel Patch position
for(auto point : anchorPatch_real) for(auto point : anchorPatch_real)
{
if(point.x - n < 0 || point.x + n >= cam.resolution(0) || point.y - n < 0 || point.y + n >= cam.resolution(1)) if(point.x - n < 0 || point.x + n >= cam.resolution(0) || point.y - n < 0 || point.y + n >= cam.resolution(1))
return false; return false;
}
for(auto point : anchorPatch_real) for(auto point : anchorPatch_real)
anchorPatch.push_back(PixelIrradiance(point, anchorImage)); anchorPatch.push_back(PixelIrradiance(point, anchorImage));
// project patch pixel to 3D space in camera coordinate system // project patch pixel to 3D space in camera coordinate system
for(auto point : und_v) for(auto point : anchorPatch_ideal)
{ anchorPatch_3d.push_back(AnchorPixelToPosition(point, cam));
anchorPatch_ideal.push_back(point);
anchorPatch_3d.push_back(projectPixelToPosition(point, cam));
}
is_anchored = true; is_anchored = true;
return true; return true;
} }
bool Feature::initializeRho(const CamStateServer& cam_states) {
// Organize camera poses and feature observations properly.
std::vector<Eigen::Isometry3d,
Eigen::aligned_allocator<Eigen::Isometry3d> > cam_poses(0);
std::vector<Eigen::Vector2d,
Eigen::aligned_allocator<Eigen::Vector2d> > measurements(0);
for (auto& m : observations) {
auto cam_state_iter = cam_states.find(m.first);
if (cam_state_iter == cam_states.end()) continue;
// Add the measurement.
measurements.push_back(m.second.head<2>());
measurements.push_back(m.second.tail<2>());
// This camera pose will take a vector from this camera frame
// to the world frame.
Eigen::Isometry3d cam0_pose;
cam0_pose.linear() = quaternionToRotation(
cam_state_iter->second.orientation).transpose();
cam0_pose.translation() = cam_state_iter->second.position;
Eigen::Isometry3d cam1_pose;
cam1_pose = cam0_pose * CAMState::T_cam0_cam1.inverse();
cam_poses.push_back(cam0_pose);
cam_poses.push_back(cam1_pose);
}
// All camera poses should be modified such that it takes a
// vector from the first camera frame in the buffer to this
// camera frame.
Eigen::Isometry3d T_c0_w = cam_poses[0];
T_anchor_w = T_c0_w;
for (auto& pose : cam_poses)
pose = pose.inverse() * T_c0_w;
// Generate initial guess
double initial_depth = 0;
initial_depth = generateInitialDepth(cam_poses[cam_poses.size()-1], measurements[0],
measurements[measurements.size()-1]);
double solution = 1.0/initial_depth;
// Apply Levenberg-Marquart method to solve for the 3d position.
double lambda = optimization_config.initial_damping;
int inner_loop_cntr = 0;
int outer_loop_cntr = 0;
bool is_cost_reduced = false;
double delta_norm = 0;
// Compute the initial cost.
double total_cost = 0.0;
for (int i = 0; i < cam_poses.size(); ++i) {
double this_cost = 0.0;
Rhocost(cam_poses[i], solution, measurements[0], measurements[i], this_cost);
total_cost += this_cost;
}
// Outer loop.
do {
Eigen::Matrix<double, 1, 1> A = Eigen::Matrix<double, 1, 1>::Zero();
Eigen::Matrix<double, 1, 1> b = Eigen::Matrix<double, 1, 1>::Zero();
for (int i = 0; i < cam_poses.size(); ++i) {
Eigen::Matrix<double, 2, 1> J;
Eigen::Vector2d r;
double w;
RhoJacobian(cam_poses[i], solution, measurements[0], measurements[i], J, r, w);
if (w == 1) {
A += J.transpose() * J;
b += J.transpose() * r;
} else {
double w_square = w * w;
A += w_square * J.transpose() * J;
b += w_square * J.transpose() * r;
}
}
// Inner loop.
// Solve for the delta that can reduce the total cost.
do {
Eigen::Matrix<double, 1, 1> damper = lambda*Eigen::Matrix<double, 1, 1>::Identity();
Eigen::Matrix<double, 1, 1> delta = (A+damper).ldlt().solve(b);
double new_solution = solution - delta(0,0);
delta_norm = delta.norm();
double new_cost = 0.0;
for (int i = 0; i < cam_poses.size(); ++i) {
double this_cost = 0.0;
Rhocost(cam_poses[i], new_solution, measurements[0], measurements[i], this_cost);
new_cost += this_cost;
}
if (new_cost < total_cost) {
is_cost_reduced = true;
solution = new_solution;
total_cost = new_cost;
lambda = lambda/10 > 1e-10 ? lambda/10 : 1e-10;
} else {
is_cost_reduced = false;
lambda = lambda*10 < 1e12 ? lambda*10 : 1e12;
}
} while (inner_loop_cntr++ <
optimization_config.inner_loop_max_iteration && !is_cost_reduced);
inner_loop_cntr = 0;
} while (outer_loop_cntr++ <
optimization_config.outer_loop_max_iteration &&
delta_norm > optimization_config.estimation_precision);
// Covert the feature position from inverse depth
// representation to its 3d coordinate.
Eigen::Vector3d final_position(measurements[0](0)/solution,
measurements[0](1)/solution, 1.0/solution);
// Check if the solution is valid. Make sure the feature
// is in front of every camera frame observing it.
bool is_valid_solution = true;
for (const auto& pose : cam_poses) {
Eigen::Vector3d position =
pose.linear()*final_position + pose.translation();
if (position(2) <= 0) {
is_valid_solution = false;
break;
}
}
//save inverse depth distance from camera
anchor_rho = solution;
// Convert the feature position to the world frame.
position = T_c0_w.linear()*final_position + T_c0_w.translation();
if (is_valid_solution)
is_initialized = true;
return is_valid_solution;
}
bool Feature::initializePosition(const CamStateServer& cam_states) { bool Feature::initializePosition(const CamStateServer& cam_states) {
// Organize camera poses and feature observations properly. // Organize camera poses and feature observations properly.
std::vector<Eigen::Isometry3d, std::vector<Eigen::Isometry3d,
Eigen::aligned_allocator<Eigen::Isometry3d> > cam_poses(0); Eigen::aligned_allocator<Eigen::Isometry3d> > cam_poses(0);
std::vector<Eigen::Vector2d, std::vector<Eigen::Vector2d,
@ -927,6 +1274,7 @@ bool Feature::initializePosition(const CamStateServer& cam_states) {
//save inverse depth distance from camera //save inverse depth distance from camera
anchor_rho = solution(2); anchor_rho = solution(2);
std::cout << "from feature: " << anchor_rho << std::endl;
// Convert the feature position to the world frame. // Convert the feature position to the world frame.
position = T_c0_w.linear()*final_position + T_c0_w.translation(); position = T_c0_w.linear()*final_position + T_c0_w.translation();

View File

@ -36,6 +36,15 @@ cv::Point2f distortPoint(
const cv::Vec4d& intrinsics, const cv::Vec4d& intrinsics,
const std::string& distortion_model, const std::string& distortion_model,
const cv::Vec4d& distortion_coeffs); const cv::Vec4d& distortion_coeffs);
void undistortPoint(
const cv::Point2f& pt_in,
const cv::Vec4d& intrinsics,
const std::string& distortion_model,
const cv::Vec4d& distortion_coeffs,
cv::Point2f& pt_out,
const cv::Matx33d &rectification_matrix = cv::Matx33d::eye(),
const cv::Vec4d &new_intrinsics = cv::Vec4d(1,1,0,0));
} }
} }
#endif #endif

View File

@ -195,9 +195,9 @@ class MsckfVio {
// for a single feature observed at a single camera frame. // for a single feature observed at a single camera frame.
void measurementJacobian(const StateIDType& cam_state_id, void measurementJacobian(const StateIDType& cam_state_id,
const FeatureIDType& feature_id, const FeatureIDType& feature_id,
Eigen::Matrix<double, 4, 6>& H_x, Eigen::Matrix<double, 2, 6>& H_x,
Eigen::Matrix<double, 4, 3>& H_f, Eigen::Matrix<double, 2, 3>& H_f,
Eigen::Vector4d& r); Eigen::Vector2d& r);
// This function computes the Jacobian of all measurements viewed // This function computes the Jacobian of all measurements viewed
// in the given camera states of this feature. // in the given camera states of this feature.
void featureJacobian(const FeatureIDType& feature_id, void featureJacobian(const FeatureIDType& feature_id,

View File

@ -22,7 +22,7 @@
<param name="PHOTOMETRIC" value="true"/> <param name="PHOTOMETRIC" value="true"/>
<!-- Debugging Flaggs --> <!-- Debugging Flaggs -->
<param name="PrintImages" value="false"/> <param name="PrintImages" value="true"/>
<param name="GroundTruth" value="false"/> <param name="GroundTruth" value="false"/>
<param name="patch_size_n" value="7"/> <param name="patch_size_n" value="7"/>

View File

@ -21,10 +21,10 @@
<param name="PHOTOMETRIC" value="true"/> <param name="PHOTOMETRIC" value="true"/>
<!-- Debugging Flaggs --> <!-- Debugging Flaggs -->
<param name="PrintImages" value="false"/> <param name="PrintImages" value="true"/>
<param name="GroundTruth" value="false"/> <param name="GroundTruth" value="false"/>
<param name="patch_size_n" value="7"/> <param name="patch_size_n" value="1"/>
<!-- Calibration parameters --> <!-- Calibration parameters -->
<rosparam command="load" file="$(arg calibration_file)"/> <rosparam command="load" file="$(arg calibration_file)"/>

View File

@ -14,6 +14,48 @@ namespace msckf_vio {
namespace image_handler { namespace image_handler {
void undistortPoint(
const cv::Point2f& pt_in,
const cv::Vec4d& intrinsics,
const std::string& distortion_model,
const cv::Vec4d& distortion_coeffs,
cv::Point2f& pt_out,
const cv::Matx33d &rectification_matrix,
const cv::Vec4d &new_intrinsics) {
std::vector<cv::Point2f> pts_in;
std::vector<cv::Point2f> pts_out;
pts_in.push_back(pt_in);
if (pts_in.size() == 0) return;
const cv::Matx33d K(
intrinsics[0], 0.0, intrinsics[2],
0.0, intrinsics[1], intrinsics[3],
0.0, 0.0, 1.0);
const cv::Matx33d K_new(
new_intrinsics[0], 0.0, new_intrinsics[2],
0.0, new_intrinsics[1], new_intrinsics[3],
0.0, 0.0, 1.0);
if (distortion_model == "radtan") {
cv::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
rectification_matrix, K_new);
} else if (distortion_model == "equidistant") {
cv::fisheye::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
rectification_matrix, K_new);
} else {
ROS_WARN_ONCE("The model %s is unrecognized, use radtan instead...",
distortion_model.c_str());
cv::undistortPoints(pts_in, pts_out, K, distortion_coeffs,
rectification_matrix, K_new);
}
pt_out = pts_out[0];
return;
}
void undistortPoints( void undistortPoints(
const std::vector<cv::Point2f>& pts_in, const std::vector<cv::Point2f>& pts_in,
const cv::Vec4d& intrinsics, const cv::Vec4d& intrinsics,

View File

@ -404,8 +404,18 @@ void MsckfVio::imageCallback(
const sensor_msgs::ImageConstPtr& cam1_img, const sensor_msgs::ImageConstPtr& cam1_img,
const CameraMeasurementConstPtr& feature_msg) const CameraMeasurementConstPtr& feature_msg)
{ {
if(PRINTIMAGES)
{
std::cout << "stopped playpack" << std::endl;
nh.setParam("/play_bag", false);
}
// Return if the gravity vector has not been set. // Return if the gravity vector has not been set.
if (!is_gravity_set) return; if (!is_gravity_set)
{
nh.setParam("/play_bag", true);
return;
}
// Start the system if the first image is received. // Start the system if the first image is received.
// The frame where the first image is received will be // The frame where the first image is received will be
@ -430,7 +440,7 @@ void MsckfVio::imageCallback(
//if(ErrState) //if(ErrState)
//{ //{
batchTruthProcessing(feature_msg->header.stamp.toSec()); //batchTruthProcessing(feature_msg->header.stamp.toSec());
if(GROUNDTRUTH) if(GROUNDTRUTH)
{ {
@ -454,7 +464,7 @@ void MsckfVio::imageCallback(
PhotometricStateAugmentation(feature_msg->header.stamp.toSec()); PhotometricStateAugmentation(feature_msg->header.stamp.toSec());
} }
else else
stateAugmentation(feature_msg->header.stamp.toSec()); PhotometricStateAugmentation(feature_msg->header.stamp.toSec());
double state_augmentation_time = ( double state_augmentation_time = (
ros::Time::now()-start_time).toSec(); ros::Time::now()-start_time).toSec();
@ -512,6 +522,13 @@ void MsckfVio::imageCallback(
publish_time, publish_time/processing_time); publish_time, publish_time/processing_time);
} }
if(PRINTIMAGES)
{
std::cout << "stopped playpack" << std::endl;
nh.setParam("/play_bag", true);
}
return; return;
} }
@ -1175,7 +1192,7 @@ void MsckfVio::PhotometricStateAugmentation(const double& time)
J * P11 * J.transpose(); J * P11 * J.transpose();
// Add photometry P_eta and surrounding Zeros // Add photometry P_eta and surrounding Zeros
state_server.state_cov(old_rows+6, old_cols+6) = irradiance_frame_bias; state_server.state_cov(old_rows+6, old_cols+6) = 0;
// Fix the covariance to be symmetric // Fix the covariance to be symmetric
MatrixXd state_cov_fixed = (state_server.state_cov + MatrixXd state_cov_fixed = (state_server.state_cov +
@ -1232,17 +1249,10 @@ void MsckfVio::PhotometricMeasurementJacobian(
Matrix3d R_w_c0 = quaternionToRotation(cam_state.orientation); Matrix3d R_w_c0 = quaternionToRotation(cam_state.orientation);
const Vector3d& t_c0_w = cam_state.position; const Vector3d& t_c0_w = cam_state.position;
// Cam1 pose.
Matrix3d R_c0_c1 = CAMState::T_cam0_cam1.linear();
Matrix3d R_w_c1 = CAMState::T_cam0_cam1.linear() * R_w_c0;
Vector3d t_c1_w = t_c0_w - R_w_c1.transpose()*CAMState::T_cam0_cam1.translation();
//photometric observation //photometric observation
std::vector<double> photo_z; std::vector<double> photo_z;
// individual Jacobians // individual Jacobians
Matrix<double, 1, 2> dI_dhj = Matrix<double, 1, 2>::Zero();
Matrix<double, 2, 3> dh_dCpij = Matrix<double, 2, 3>::Zero(); Matrix<double, 2, 3> dh_dCpij = Matrix<double, 2, 3>::Zero();
Matrix<double, 2, 3> dh_dGpij = Matrix<double, 2, 3>::Zero(); Matrix<double, 2, 3> dh_dGpij = Matrix<double, 2, 3>::Zero();
Matrix<double, 2, 6> dh_dXplj = Matrix<double, 2, 6>::Zero(); Matrix<double, 2, 6> dh_dXplj = Matrix<double, 2, 6>::Zero();
@ -1254,36 +1264,19 @@ void MsckfVio::PhotometricMeasurementJacobian(
Matrix<double, 3, 3> dCpij_dGpC = Matrix<double, 3, 3>::Zero(); Matrix<double, 3, 3> dCpij_dGpC = Matrix<double, 3, 3>::Zero();
// one line of the NxN Jacobians // one line of the NxN Jacobians
Eigen::Matrix<double, 1, 1> H_rhoj; Eigen::Matrix<double, 2, 1> H_rho;
Eigen::Matrix<double, 1, 6> H_plj; Eigen::Matrix<double, 2, 6> H_plj;
Eigen::Matrix<double, 1, 6> H_pAj; Eigen::Matrix<double, 2, 6> H_pAj;
// combined Jacobians
Eigen::MatrixXd H_rho(N*N, 1);
Eigen::MatrixXd H_pl(N*N, 6);
Eigen::MatrixXd H_pA(N*N, 6);
auto frame = cam0.moving_window.find(cam_state_id)->second.image; auto frame = cam0.moving_window.find(cam_state_id)->second.image;
int count = 0; int count = 0;
double dx, dy;
for (auto point : feature.anchorPatch_3d) auto point = feature.anchorPatch_3d[0];
{
Eigen::Vector3d p_c0 = R_w_c0 * (point-t_c0_w); Eigen::Vector3d p_c0 = R_w_c0 * (point-t_c0_w);
cv::Point2f p_in_c0 = feature.projectPositionToCamera(cam_state, cam_state_id, cam0, point);
//add observation
photo_z.push_back(feature.PixelIrradiance(p_in_c0, frame));
// add jacobian // add jacobian
// frame derivative calculated convoluting with kernel [-1, 0, 1]
dx = feature.PixelIrradiance(cv::Point2f(p_in_c0.x+1, p_in_c0.y), frame) - feature.PixelIrradiance(cv::Point2f(p_in_c0.x-1, p_in_c0.y), frame);
dy = feature.PixelIrradiance(cv::Point2f(p_in_c0.x, p_in_c0.y+1), frame) - feature.PixelIrradiance(cv::Point2f(p_in_c0.x, p_in_c0.y-1), frame);
dI_dhj(0, 0) = dx;
dI_dhj(0, 1) = dy;
//dh / d{}^Cp_{ij} //dh / d{}^Cp_{ij}
dh_dCpij(0, 0) = 1 / p_c0(2); dh_dCpij(0, 0) = 1 / p_c0(2);
dh_dCpij(1, 1) = 1 / p_c0(2); dh_dCpij(1, 1) = 1 / p_c0(2);
@ -1292,7 +1285,7 @@ void MsckfVio::PhotometricMeasurementJacobian(
dCpij_dGpij = quaternionToRotation(cam_state.orientation); dCpij_dGpij = quaternionToRotation(cam_state.orientation);
//orientation takes camera frame to world frame, we wa //orientation takes camera frame to world frame
dh_dGpij = dh_dCpij * dCpij_dGpij; dh_dGpij = dh_dCpij * dCpij_dGpij;
//dh / d X_{pl} //dh / d X_{pl}
@ -1306,6 +1299,10 @@ void MsckfVio::PhotometricMeasurementJacobian(
// Isometry T_anchor_w takes a vector in anchor frame to world frame // Isometry T_anchor_w takes a vector in anchor frame to world frame
dGpj_drhoj = -feature.T_anchor_w.linear() * Eigen::Vector3d(feature.anchorPatch_ideal[count].x/(rho*rho), feature.anchorPatch_ideal[count].y/(rho*rho), 1/(rho*rho)); dGpj_drhoj = -feature.T_anchor_w.linear() * Eigen::Vector3d(feature.anchorPatch_ideal[count].x/(rho*rho), feature.anchorPatch_ideal[count].y/(rho*rho), 1/(rho*rho));
// alternative derivation towards feature
Matrix3d dCpc0_dpg = R_w_c0;
dGpj_XpAj.block<3, 3>(0, 0) = - feature.T_anchor_w.linear() dGpj_XpAj.block<3, 3>(0, 0) = - feature.T_anchor_w.linear()
* skewSymmetric(Eigen::Vector3d(feature.anchorPatch_ideal[count].x/(rho), * skewSymmetric(Eigen::Vector3d(feature.anchorPatch_ideal[count].x/(rho),
feature.anchorPatch_ideal[count].y/(rho), feature.anchorPatch_ideal[count].y/(rho),
@ -1313,48 +1310,31 @@ void MsckfVio::PhotometricMeasurementJacobian(
dGpj_XpAj.block<3, 3>(0, 3) = Matrix<double, 3, 3>::Identity(); dGpj_XpAj.block<3, 3>(0, 3) = Matrix<double, 3, 3>::Identity();
// Intermediate Jakobians // Intermediate Jakobians
H_rhoj = dI_dhj * dh_dGpij * dGpj_drhoj; // 1 x 1 H_rho = dh_dGpij * dGpj_drhoj; // 2 x 1
H_plj = dI_dhj * dh_dXplj; // 1 x 6 H_plj = dh_dXplj; // 2 x 6
H_pAj = dI_dhj * dh_dGpij * dGpj_XpAj; // 1 x 6 H_pAj = dh_dGpij * dGpj_XpAj; // 2 x 6
H_rho.block<1, 1>(count, 0) = H_rhoj;
H_pl.block<1, 6>(count, 0) = H_plj;
H_pA.block<1, 6>(count, 0) = H_pAj;
count++;
}
// calculate residual // calculate residual
//observation //observation
const Vector4d& z = feature.observations.find(cam_state_id)->second; const Vector4d& total_z = feature.observations.find(cam_state_id)->second;
const Vector2d z = Vector2d(total_z[0], total_z[1]);
//estimate photometric measurement VectorXd r_i = VectorXd::Zero(2);
std::vector<double> estimate_irradiance;
std::vector<double> estimate_photo_z;
IlluminationParameter estimated_illumination;
feature.estimate_FrameIrradiance(cam_state, cam_state_id, cam0, estimate_irradiance, estimated_illumination);
// calculated here, because we need true 'estimate_irradiance' later for jacobi //calculate residual
for (auto& estimate_irradiance_j : estimate_irradiance)
estimate_photo_z.push_back (estimate_irradiance_j *
estimated_illumination.frame_gain * estimated_illumination.feature_gain +
estimated_illumination.frame_bias + estimated_illumination.feature_bias);
std::vector<double> photo_r; r_i[0] = z[0] - p_c0(0)/p_c0(2);
r_i[1] = z[1] - p_c0(1)/p_c0(2);
//calculate photom. residual MatrixXd H_xl = MatrixXd::Zero(2, 21+state_server.cam_states.size()*7);
for(int i = 0; i < photo_z.size(); i++)
photo_r.push_back(photo_z[i] - estimate_photo_z[i]);
MatrixXd H_xl = MatrixXd::Zero(N*N, 21+state_server.cam_states.size()*7);
MatrixXd H_yl = MatrixXd::Zero(N*N, N*N+state_server.cam_states.size()+1);
// set anchor Jakobi // set anchor Jakobi
// get position of anchor in cam states // get position of anchor in cam states
auto cam_state_anchor = state_server.cam_states.find(feature.observations.begin()->first); auto cam_state_anchor = state_server.cam_states.find(feature.observations.begin()->first);
int cam_state_cntr_anchor = std::distance(state_server.cam_states.begin(), cam_state_anchor); int cam_state_cntr_anchor = std::distance(state_server.cam_states.begin(), cam_state_anchor);
H_xl.block(0, 21+cam_state_cntr_anchor*7, N*N, 6) = -H_pA; H_xl.block(0, 21+cam_state_cntr_anchor*7, 2, 6) = H_pAj;
// set frame Jakobi // set frame Jakobi
//get position of current frame in cam states //get position of current frame in cam states
@ -1362,32 +1342,19 @@ void MsckfVio::PhotometricMeasurementJacobian(
int cam_state_cntr = std::distance(state_server.cam_states.begin(), cam_state_iter); int cam_state_cntr = std::distance(state_server.cam_states.begin(), cam_state_iter);
// set jakobi of state // set jakobi of state
H_xl.block(0, 21+cam_state_cntr*7, N*N, 6) = -H_pl; H_xl.block(0, 21+cam_state_cntr*7, 2, 6) = H_plj;
// set ones for irradiance bias
H_xl.block(0, 21+cam_state_cntr*7+6, N*N, 1) = Eigen::ArrayXd::Ones(N*N);
// set irradiance error Block
H_yl.block(0, 0,N*N, N*N) = estimated_illumination.feature_gain * estimated_illumination.frame_gain * Eigen::MatrixXd::Identity(N*N, N*N);
// TODO make this calculation more fluent
for(int i = 0; i< N*N; i++)
H_yl(i, N*N+cam_state_cntr) = estimate_irradiance[i];
H_yl.block(0, N*N+state_server.cam_states.size(), N*N, 1) = -H_rho;
H_x = H_xl; H_x = H_xl;
H_y = H_yl; H_y = H_rho;
r = r_i;
//TODO make this more fluent as well //TODO make this more fluent as well
count = 0;
for(auto data : photo_r)
r[count++] = data;
std::stringstream ss;
ss << "INFO:" << " anchor: " << cam_state_cntr_anchor << " frame: " << cam_state_cntr;
if(PRINTIMAGES) if(PRINTIMAGES)
{ {
feature.MarkerGeneration(marker_pub, state_server.cam_states); //std::stringstream ss;
feature.VisualizePatch(cam_state, cam_state_id, cam0, photo_r, ss); //ss << "INFO:" << " anchor: " << cam_state_cntr_anchor << " frame: " << cam_state_cntr;
//feature.MarkerGeneration(marker_pub, state_server.cam_states);
//feature.VisualizePatch(cam_state, cam_state_id, cam0, photo_r, ss);
} }
return; return;
@ -1400,28 +1367,6 @@ void MsckfVio::PhotometricFeatureJacobian(
{ {
// stop playing bagfile if printing images // stop playing bagfile if printing images
if(PRINTIMAGES)
{
std::cout << "stopped playpack" << std::endl;
nh.setParam("/play_bag", false);
}
// Errstate
calcErrorState();
/*
std::cout << "--- error state: ---\n " << std::endl;
std::cout << "imu orientation:\n " << err_state_server.imu_state.orientation << std::endl;
std::cout << "imu position\n" << err_state_server.imu_state.position << std::endl;
int count = 0;
for(auto cam_state : err_state_server.cam_states)
{
std::cout << " - cam " << count++ << " - \n" << std::endl;
std::cout << "orientation: " << cam_state.second.orientation(0) << cam_state.second.orientation(1) << " " << cam_state.second.orientation(2) << " " << std::endl;
std::cout << "position:" << cam_state.second.position(0) << " " << cam_state.second.position(1) << " " << cam_state.second.position(2) << std::endl;
}
*/
const auto& feature = map_server[feature_id]; const auto& feature = map_server[feature_id];
@ -1433,15 +1378,17 @@ void MsckfVio::PhotometricFeatureJacobian(
if (feature.observations.find(cam_id) == if (feature.observations.find(cam_id) ==
feature.observations.end()) continue; feature.observations.end()) continue;
if (feature.observations.find(cam_id) ==
feature.observations.begin()) continue;
valid_cam_state_ids.push_back(cam_id); valid_cam_state_ids.push_back(cam_id);
} }
int jacobian_row_size = 0; int jacobian_row_size = 0;
jacobian_row_size = N * N * valid_cam_state_ids.size(); jacobian_row_size = 2 * valid_cam_state_ids.size();
MatrixXd H_xi = MatrixXd::Zero(jacobian_row_size, MatrixXd H_xi = MatrixXd::Zero(jacobian_row_size,
21+state_server.cam_states.size()*7); 21+state_server.cam_states.size()*7);
MatrixXd H_yi = MatrixXd::Zero(jacobian_row_size, N*N+state_server.cam_states.size()+1); MatrixXd H_yi = MatrixXd::Zero(jacobian_row_size, 1);
VectorXd r_i = VectorXd::Zero(jacobian_row_size); VectorXd r_i = VectorXd::Zero(jacobian_row_size);
int stack_cntr = 0; int stack_cntr = 0;
@ -1449,10 +1396,9 @@ void MsckfVio::PhotometricFeatureJacobian(
MatrixXd H_xl; MatrixXd H_xl;
MatrixXd H_yl; MatrixXd H_yl;
Eigen::VectorXd r_l = VectorXd::Zero(N*N); Eigen::VectorXd r_l = VectorXd::Zero(2);
PhotometricMeasurementJacobian(cam_id, feature.id, H_xl, H_yl, r_l); PhotometricMeasurementJacobian(cam_id, feature.id, H_xl, H_yl, r_l);
auto cam_state_iter = state_server.cam_states.find(cam_id); auto cam_state_iter = state_server.cam_states.find(cam_id);
int cam_state_cntr = std::distance( int cam_state_cntr = std::distance(
state_server.cam_states.begin(), cam_state_iter); state_server.cam_states.begin(), cam_state_iter);
@ -1460,8 +1406,8 @@ void MsckfVio::PhotometricFeatureJacobian(
// Stack the Jacobians. // Stack the Jacobians.
H_xi.block(stack_cntr, 0, H_xl.rows(), H_xl.cols()) = H_xl; H_xi.block(stack_cntr, 0, H_xl.rows(), H_xl.cols()) = H_xl;
H_yi.block(stack_cntr, 0, H_yl.rows(), H_yl.cols()) = H_yl; H_yi.block(stack_cntr, 0, H_yl.rows(), H_yl.cols()) = H_yl;
r_i.segment(stack_cntr, N*N) = r_l; r_i.segment(stack_cntr, 2) = r_l;
stack_cntr += N*N; stack_cntr += 2;
} }
// Project the residual and Jacobians onto the nullspace // Project the residual and Jacobians onto the nullspace
@ -1470,32 +1416,47 @@ void MsckfVio::PhotometricFeatureJacobian(
// get Nullspace // get Nullspace
FullPivLU<MatrixXd> lu(H_yi.transpose()); FullPivLU<MatrixXd> lu(H_yi.transpose());
MatrixXd A_null_space = lu.kernel(); MatrixXd A_null_space = lu.kernel();
/*
JacobiSVD<MatrixXd> svd_helper(H_yi, ComputeFullU | ComputeThinV);
int sv_size = 0;
Eigen::VectorXd singularValues = svd_helper.singularValues();
for(int i = 0; i < singularValues.size(); i++)
if(singularValues[i] > 1e-12)
sv_size++;
MatrixXd A = svd_helper.matrixU().rightCols(jacobian_row_size - singularValues.size());
*/
H_x = A_null_space.transpose() * H_xi; H_x = A_null_space.transpose() * H_xi;
r = A_null_space.transpose() * r_i; r = A_null_space.transpose() * r_i;
/*
if(PRINTIMAGES) if(PRINTIMAGES)
{ {
ofstream myfile;
myfile.open("/home/raphael/dev/octave/log2octave");
myfile << "# Created by Octave 3.8.1, Wed Jun 12 14:36:37 2019 CEST <raphael@raphael-desktop>\n"
<< "# name: Hx\n"
<< "# type: matrix\n"
<< "# rows: " << H_xi.rows() << "\n"
<< "# columns: " << H_xi.cols() << "\n"
<< H_xi << endl;
myfile << "# name: Hy\n"
<< "# type: matrix\n"
<< "# rows: " << H_yi.rows() << "\n"
<< "# columns: " << H_yi.cols() << "\n"
<< H_yi << endl;
myfile << "# name: r\n"
<< "# type: matrix\n"
<< "# rows: " << r_i.rows() << "\n"
<< "# columns: " << 1 << "\n"
<< r_i << endl;
myfile.close();
std::cout << "resume playback" << std::endl; std::cout << "resume playback" << std::endl;
nh.setParam("/play_bag", true); nh.setParam("/play_bag", true);
} }*/
return; return;
} }
void MsckfVio::measurementJacobian( void MsckfVio::measurementJacobian(
const StateIDType& cam_state_id, const StateIDType& cam_state_id,
const FeatureIDType& feature_id, const FeatureIDType& feature_id,
Matrix<double, 4, 6>& H_x, Matrix<double, 4, 3>& H_f, Vector4d& r) Matrix<double, 2, 6>& H_x, Matrix<double, 2, 3>& H_f, Vector2d& r)
{ {
// Prepare all the required data. // Prepare all the required data.
@ -1514,48 +1475,42 @@ void MsckfVio::measurementJacobian(
// 3d feature position in the world frame. // 3d feature position in the world frame.
// And its observation with the stereo cameras. // And its observation with the stereo cameras.
const Vector3d& p_w = feature.position; const Vector3d& p_w = feature.position;
const Vector4d& z = feature.observations.find(cam_state_id)->second; const Vector2d& z = feature.observations.find(cam_state_id)->second.topRows(2);
// Convert the feature position from the world frame to // Convert the feature position from the world frame to
// the cam0 and cam1 frame. // the cam0 and cam1 frame.
Vector3d p_c0 = R_w_c0 * (p_w-t_c0_w); Vector3d p_c0 = R_w_c0 * (p_w-t_c0_w);
Vector3d p_c1 = R_w_c1 * (p_w-t_c1_w); //Vector3d p_c1 = R_w_c1 * (p_w-t_c1_w);
// Compute the Jacobians. // Compute the Jacobians.
Matrix<double, 4, 3> dz_dpc0 = Matrix<double, 4, 3>::Zero(); Matrix<double, 2, 3> dz_dpc0 = Matrix<double, 2, 3>::Zero();
dz_dpc0(0, 0) = 1 / p_c0(2); dz_dpc0(0, 0) = 1 / p_c0(2);
dz_dpc0(1, 1) = 1 / p_c0(2); dz_dpc0(1, 1) = 1 / p_c0(2);
dz_dpc0(0, 2) = -p_c0(0) / (p_c0(2)*p_c0(2)); dz_dpc0(0, 2) = -p_c0(0) / (p_c0(2)*p_c0(2));
dz_dpc0(1, 2) = -p_c0(1) / (p_c0(2)*p_c0(2)); dz_dpc0(1, 2) = -p_c0(1) / (p_c0(2)*p_c0(2));
/*
Matrix<double, 4, 3> dz_dpc1 = Matrix<double, 4, 3>::Zero(); Matrix<double, 4, 3> dz_dpc1 = Matrix<double, 4, 3>::Zero();
dz_dpc1(2, 0) = 1 / p_c1(2); dz_dpc1(2, 0) = 1 / p_c1(2);
dz_dpc1(3, 1) = 1 / p_c1(2); dz_dpc1(3, 1) = 1 / p_c1(2);
dz_dpc1(2, 2) = -p_c1(0) / (p_c1(2)*p_c1(2)); dz_dpc1(2, 2) = -p_c1(0) / (p_c1(2)*p_c1(2));
dz_dpc1(3, 2) = -p_c1(1) / (p_c1(2)*p_c1(2)); dz_dpc1(3, 2) = -p_c1(1) / (p_c1(2)*p_c1(2));
*/
Matrix<double, 3, 6> dpc0_dxc = Matrix<double, 3, 6>::Zero(); Matrix<double, 3, 6> dpc0_dxc = Matrix<double, 3, 6>::Zero();
// original jacobi // original jacobi
//dpc0_dxc.leftCols(3) = skewSymmetric(p_c0);
// my version of calculation
dpc0_dxc.leftCols(3) = skewSymmetric(p_c0); dpc0_dxc.leftCols(3) = skewSymmetric(p_c0);
//dpc0_dxc.leftCols(3) = - skewSymmetric(R_w_c0.transpose() * (t_c0_w - p_w)) * R_w_c0;
dpc0_dxc.rightCols(3) = -R_w_c0; dpc0_dxc.rightCols(3) = -R_w_c0;
Matrix<double, 3, 6> dpc1_dxc = Matrix<double, 3, 6>::Zero();
dpc1_dxc.leftCols(3) = R_c0_c1 * skewSymmetric(p_c0);
dpc1_dxc.rightCols(3) = -R_w_c1;
Matrix3d dpc0_dpg = R_w_c0; Matrix3d dpc0_dpg = R_w_c0;
Matrix3d dpc1_dpg = R_w_c1; Matrix3d dpc1_dpg = R_w_c1;
H_x = dz_dpc0*dpc0_dxc + dz_dpc1*dpc1_dxc; H_x = dz_dpc0*dpc0_dxc; //+ dz_dpc1*dpc1_dxc;
H_f = dz_dpc0*dpc0_dpg + dz_dpc1*dpc1_dpg; H_f = dz_dpc0*dpc0_dpg; // + dz_dpc1*dpc1_dpg;
// Compute the residual. // Compute the residual.
r = z - Vector4d(p_c0(0)/p_c0(2), p_c0(1)/p_c0(2), r = z - Vector2d(p_c0(0)/p_c0(2), p_c0(1)/p_c0(2));//,
p_c1(0)/p_c1(2), p_c1(1)/p_c1(2)); //p_c1(0)/p_c1(2), p_c1(1)/p_c1(2));
return; return;
} }
@ -1579,19 +1534,19 @@ void MsckfVio::featureJacobian(
} }
int jacobian_row_size = 0; int jacobian_row_size = 0;
jacobian_row_size = 4 * valid_cam_state_ids.size(); jacobian_row_size = 2 * valid_cam_state_ids.size();
MatrixXd H_xj = MatrixXd::Zero(jacobian_row_size, MatrixXd H_xj = MatrixXd::Zero(jacobian_row_size,
21+state_server.cam_states.size()*6); 21+state_server.cam_states.size()*7);
MatrixXd H_fj = MatrixXd::Zero(jacobian_row_size, 3); MatrixXd H_fj = MatrixXd::Zero(jacobian_row_size, 3);
VectorXd r_j = VectorXd::Zero(jacobian_row_size); VectorXd r_j = VectorXd::Zero(jacobian_row_size);
int stack_cntr = 0; int stack_cntr = 0;
for (const auto& cam_id : valid_cam_state_ids) { for (const auto& cam_id : valid_cam_state_ids) {
Matrix<double, 4, 6> H_xi = Matrix<double, 4, 6>::Zero(); Matrix<double, 2, 6> H_xi = Matrix<double, 2, 6>::Zero();
Matrix<double, 4, 3> H_fi = Matrix<double, 4, 3>::Zero(); Matrix<double, 2, 3> H_fi = Matrix<double, 2, 3>::Zero();
Vector4d r_i = Vector4d::Zero(); Vector2d r_i = Vector2d::Zero();
measurementJacobian(cam_id, feature.id, H_xi, H_fi, r_i); measurementJacobian(cam_id, feature.id, H_xi, H_fi, r_i);
auto cam_state_iter = state_server.cam_states.find(cam_id); auto cam_state_iter = state_server.cam_states.find(cam_id);
@ -1599,10 +1554,10 @@ void MsckfVio::featureJacobian(
state_server.cam_states.begin(), cam_state_iter); state_server.cam_states.begin(), cam_state_iter);
// Stack the Jacobians. // Stack the Jacobians.
H_xj.block<4, 6>(stack_cntr, 21+6*cam_state_cntr) = H_xi; H_xj.block<2, 6>(stack_cntr, 21+7*cam_state_cntr) = H_xi;
H_fj.block<4, 3>(stack_cntr, 0) = H_fi; H_fj.block<2, 3>(stack_cntr, 0) = H_fi;
r_j.segment<4>(stack_cntr) = r_i; r_j.segment<2>(stack_cntr) = r_i;
stack_cntr += 4; stack_cntr += 2;
} }
// Project the residual and Jacobians onto the nullspace // Project the residual and Jacobians onto the nullspace
@ -1629,12 +1584,12 @@ void MsckfVio::featureJacobian(
/* /*
ofstream myfile; ofstream myfile;
myfile.open ("/home/raphael/dev/MSCKF_ws/log.txt"); myfile.open ("/home/raphael/dev/MSCKF_ws/log.txt");
myfile << "-- residual -- \n" << r << "\n---- H ----\n" << H_x << "\n---- state cov ----\n" << state_server.state_cov <<endl; myfile << "Hx\n" << H_x << "r\n" << r << "from residual estimated error state: " << H_x.colPivHouseholderQr().solve(r) << endl;
myfile.close(); myfile.close();
cout << "---------- LOGGED -------- " << endl;
*/
nh.setParam("/play_bag", false);
cout << "---------- LOGGED -------- " << endl;
nh.setParam("/play_bag", false);
*/
return; return;
} }
@ -1646,7 +1601,7 @@ void MsckfVio::measurementUpdate(const MatrixXd& H, const VectorXd& r) {
// complexity as in Equation (28), (29). // complexity as in Equation (28), (29).
MatrixXd H_thin; MatrixXd H_thin;
VectorXd r_thin; VectorXd r_thin;
/*
if (H.rows() > H.cols()) { if (H.rows() > H.cols()) {
// Convert H to a sparse matrix. // Convert H to a sparse matrix.
SparseMatrix<double> H_sparse = H.sparseView(); SparseMatrix<double> H_sparse = H.sparseView();
@ -1661,8 +1616,8 @@ void MsckfVio::measurementUpdate(const MatrixXd& H, const VectorXd& r) {
(spqr_helper.matrixQ().transpose() * H).evalTo(H_temp); (spqr_helper.matrixQ().transpose() * H).evalTo(H_temp);
(spqr_helper.matrixQ().transpose() * r).evalTo(r_temp); (spqr_helper.matrixQ().transpose() * r).evalTo(r_temp);
H_thin = H_temp.topRows(21+state_server.cam_states.size()*6); H_thin = H_temp.topRows(21+state_server.cam_states.size()*7);
r_thin = r_temp.head(21+state_server.cam_states.size()*6); r_thin = r_temp.head(21+state_server.cam_states.size()*7);
//HouseholderQR<MatrixXd> qr_helper(H); //HouseholderQR<MatrixXd> qr_helper(H);
//MatrixXd Q = qr_helper.householderQ(); //MatrixXd Q = qr_helper.householderQ();
@ -1670,10 +1625,10 @@ void MsckfVio::measurementUpdate(const MatrixXd& H, const VectorXd& r) {
//H_thin = Q1.transpose() * H; //H_thin = Q1.transpose() * H;
//r_thin = Q1.transpose() * r; //r_thin = Q1.transpose() * r;
} else { } else {*/
H_thin = H; H_thin = H;
r_thin = r; r_thin = r;
} //}
// Compute the Kalman gain. // Compute the Kalman gain.
const MatrixXd& P = state_server.state_cov; const MatrixXd& P = state_server.state_cov;
@ -1720,7 +1675,7 @@ void MsckfVio::measurementUpdate(const MatrixXd& H, const VectorXd& r) {
auto cam_state_iter = state_server.cam_states.begin(); auto cam_state_iter = state_server.cam_states.begin();
for (int i = 0; i < state_server.cam_states.size(); for (int i = 0; i < state_server.cam_states.size();
++i, ++cam_state_iter) { ++i, ++cam_state_iter) {
const VectorXd& delta_x_cam = delta_x.segment<6>(21+i*6); const VectorXd& delta_x_cam = delta_x.segment<6>(21+i*7);
const Vector4d dq_cam = smallAngleQuaternion(delta_x_cam.head<3>()); const Vector4d dq_cam = smallAngleQuaternion(delta_x_cam.head<3>());
cam_state_iter->second.orientation = quaternionMultiplication( cam_state_iter->second.orientation = quaternionMultiplication(
dq_cam, cam_state_iter->second.orientation); dq_cam, cam_state_iter->second.orientation);
@ -1795,7 +1750,7 @@ void MsckfVio::removeLostFeatures() {
invalid_feature_ids.push_back(feature.id); invalid_feature_ids.push_back(feature.id);
continue; continue;
} else { } else {
if(!feature.initializePosition(state_server.cam_states)) { if(!feature.initializeRho(state_server.cam_states)) {
invalid_feature_ids.push_back(feature.id); invalid_feature_ids.push_back(feature.id);
continue; continue;
} }
@ -1812,9 +1767,9 @@ void MsckfVio::removeLostFeatures() {
if(PHOTOMETRIC) if(PHOTOMETRIC)
//just use max. size, as gets shrunken down after anyway //just use max. size, as gets shrunken down after anyway
jacobian_row_size += N*N*feature.observations.size(); jacobian_row_size += 2*feature.observations.size();
else else
jacobian_row_size += 4*feature.observations.size() - 3; jacobian_row_size += 2*feature.observations.size() - 3;
processed_feature_ids.push_back(feature.id); processed_feature_ids.push_back(feature.id);
} }
@ -1831,7 +1786,7 @@ void MsckfVio::removeLostFeatures() {
if (processed_feature_ids.size() == 0) return; if (processed_feature_ids.size() == 0) return;
MatrixXd H_x = MatrixXd::Zero(jacobian_row_size, MatrixXd H_x = MatrixXd::Zero(jacobian_row_size,
21+augmentationSize*state_server.cam_states.size()); 21+7*state_server.cam_states.size());
VectorXd r = VectorXd::Zero(jacobian_row_size); VectorXd r = VectorXd::Zero(jacobian_row_size);
int stack_cntr = 0; int stack_cntr = 0;
@ -1965,7 +1920,7 @@ void MsckfVio::pruneCamStateBuffer() {
feature.observations.erase(cam_id); feature.observations.erase(cam_id);
continue; continue;
} else { } else {
if(!feature.initializePosition(state_server.cam_states)) { if(!feature.initializeRho(state_server.cam_states)) {
for (const auto& cam_id : involved_cam_state_ids) for (const auto& cam_id : involved_cam_state_ids)
feature.observations.erase(cam_id); feature.observations.erase(cam_id);
continue; continue;
@ -1982,9 +1937,9 @@ void MsckfVio::pruneCamStateBuffer() {
} }
} }
if(PHOTOMETRIC) if(PHOTOMETRIC)
jacobian_row_size += N*N*involved_cam_state_ids.size(); jacobian_row_size += 2*involved_cam_state_ids.size();
else else
jacobian_row_size += 4*involved_cam_state_ids.size() - 3; jacobian_row_size += 2*involved_cam_state_ids.size() - 3;
} }
//cout << "jacobian row #: " << jacobian_row_size << endl; //cout << "jacobian row #: " << jacobian_row_size << endl;
@ -1992,7 +1947,7 @@ void MsckfVio::pruneCamStateBuffer() {
// Compute the Jacobian and residual. // Compute the Jacobian and residual.
MatrixXd H_xj; MatrixXd H_xj;
VectorXd r_j; VectorXd r_j;
MatrixXd H_x = MatrixXd::Zero(jacobian_row_size, 21+augmentationSize*state_server.cam_states.size()); MatrixXd H_x = MatrixXd::Zero(jacobian_row_size, 21+7*state_server.cam_states.size());
VectorXd r = VectorXd::Zero(jacobian_row_size); VectorXd r = VectorXd::Zero(jacobian_row_size);
int stack_cntr = 0; int stack_cntr = 0;
for (auto& item : map_server) { for (auto& item : map_server) {
@ -2037,8 +1992,8 @@ void MsckfVio::pruneCamStateBuffer() {
for (const auto& cam_id : rm_cam_state_ids) { for (const auto& cam_id : rm_cam_state_ids) {
int cam_sequence = std::distance(state_server.cam_states.begin(), int cam_sequence = std::distance(state_server.cam_states.begin(),
state_server.cam_states.find(cam_id)); state_server.cam_states.find(cam_id));
int cam_state_start = 21 + augmentationSize*cam_sequence; int cam_state_start = 21 + 7*cam_sequence;
int cam_state_end = cam_state_start + augmentationSize; int cam_state_end = cam_state_start + 7;
// Remove the corresponding rows and columns in the state // Remove the corresponding rows and columns in the state
@ -2059,10 +2014,10 @@ void MsckfVio::pruneCamStateBuffer() {
state_server.state_cov.cols()-cam_state_end); state_server.state_cov.cols()-cam_state_end);
state_server.state_cov.conservativeResize( state_server.state_cov.conservativeResize(
state_server.state_cov.rows()-augmentationSize, state_server.state_cov.cols()-augmentationSize); state_server.state_cov.rows()-7, state_server.state_cov.cols()-7);
} else { } else {
state_server.state_cov.conservativeResize( state_server.state_cov.conservativeResize(
state_server.state_cov.rows()-augmentationSize, state_server.state_cov.cols()-augmentationSize); state_server.state_cov.rows()-7, state_server.state_cov.cols()-7);
} }
// Remove this camera state in the state vector. // Remove this camera state in the state vector.