laser_scan_matcher.cpp

Go to the documentation of this file.

/*
 * Copyright (c) 2011, Ivan Dryanovski, William Morris
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 *     * Redistributions of source code must retain the above copyright
 *       notice, this list of conditions and the following disclaimer.
 *     * Redistributions in binary form must reproduce the above copyright
 *       notice, this list of conditions and the following disclaimer in the
 *       documentation and/or other materials provided with the distribution.
 *     * Neither the name of the CCNY Robotics Lab nor the names of its
 *       contributors may be used to endorse or promote products derived from
 *       this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 */

/*  This package uses Canonical Scan Matcher [1], written by
 *  Andrea Censi
 *
 *  [1] A. Censi, "An ICP variant using a point-to-line metric"
 *  Proceedings of the IEEE International Conference
 *  on Robotics and Automation (ICRA), 2008
 */

#include <laser_scan_matcher/laser_scan_matcher.h>
#include <pcl_conversions/pcl_conversions.h>
#include <boost/assign.hpp>
#include <tf/transform_datatypes.h>

namespace scan_tools
{

LaserScanMatcher::LaserScanMatcher(ros::NodeHandle nh, ros::NodeHandle nh_private):
  nh_(nh),
  nh_private_(nh_private),
  initialized_(false),
  received_imu_(false),
  received_odom_(false),
  received_vel_(false)
{
  ROS_INFO("Starting LaserScanMatcher");

  // **** init parameters

  initParams();

  // **** state variables

  last_base_in_fixed_.setIdentity();
  keyframe_base_in_fixed_.setIdentity();
  last_used_odom_pose_.setIdentity();
  input_.laser[0] = 0.0;
  input_.laser[1] = 0.0;
  input_.laser[2] = 0.0;

  // Initialize output_ vectors as Null for error-checking
  output_.cov_x_m = 0;
  output_.dx_dy1_m = 0;
  output_.dx_dy2_m = 0;

  // **** publishers

  if (publish_pose_)
  {
    pose_publisher_  = nh_.advertise<geometry_msgs::Pose2D>(
      "pose2D", 5);
  }

  if (publish_pose_stamped_)
  {
    pose_stamped_publisher_ = nh_.advertise<geometry_msgs::PoseStamped>(
      "pose_stamped", 5);
  }

  if (publish_pose_with_covariance_)
  {
    pose_with_covariance_publisher_  = nh_.advertise<geometry_msgs::PoseWithCovariance>(
      "pose_with_covariance", 5);
  }

  if (publish_pose_with_covariance_stamped_)
  {
    pose_with_covariance_stamped_publisher_ = nh_.advertise<geometry_msgs::PoseWithCovarianceStamped>(
      "pose_with_covariance_stamped", 5);
  }

  // *** subscribers

  if (use_cloud_input_)
  {
    cloud_subscriber_ = nh_.subscribe(
      "cloud", 1, &LaserScanMatcher::cloudCallback, this);
  }
  else
  {
    scan_subscriber_ = nh_.subscribe(
      "scan", 1, &LaserScanMatcher::scanCallback, this);
  }

  if (use_imu_)
  {
    imu_subscriber_ = nh_.subscribe(
      "imu/data", 1, &LaserScanMatcher::imuCallback, this);
  }
  if (use_odom_)
  {
    odom_subscriber_ = nh_.subscribe(
      "odom", 1, &LaserScanMatcher::odomCallback, this);
  }
  if (use_vel_)
  {
    if (stamped_vel_)
      vel_subscriber_ = nh_.subscribe(
        "vel", 1, &LaserScanMatcher::velStmpCallback, this);
    else
      vel_subscriber_ = nh_.subscribe(
        "vel", 1, &LaserScanMatcher::velCallback, this);
  }
}

LaserScanMatcher::~LaserScanMatcher()
{
  ROS_INFO("Destroying LaserScanMatcher");
}

void LaserScanMatcher::initParams()
{
  if (!nh_private_.getParam ("base_frame", base_frame_))
    base_frame_ = "base_link";
  if (!nh_private_.getParam ("fixed_frame", fixed_frame_))
    fixed_frame_ = "world";

  // **** input type - laser scan, or point clouds?
  // if false, will subscribe to LaserScan msgs on /scan.
  // if true, will subscribe to PointCloud2 msgs on /cloud

  if (!nh_private_.getParam ("use_cloud_input", use_cloud_input_))
    use_cloud_input_= false;

  if (use_cloud_input_)
  {
    if (!nh_private_.getParam ("cloud_range_min", cloud_range_min_))
      cloud_range_min_ = 0.1;
    if (!nh_private_.getParam ("cloud_range_max", cloud_range_max_))
      cloud_range_max_ = 50.0;
    if (!nh_private_.getParam ("cloud_res", cloud_res_))
      cloud_res_ = 0.05;

    input_.min_reading = cloud_range_min_;
    input_.max_reading = cloud_range_max_;
  }

  // **** keyframe params: when to generate the keyframe scan
  // if either is set to 0, reduces to frame-to-frame matching

  if (!nh_private_.getParam ("kf_dist_linear", kf_dist_linear_))
    kf_dist_linear_ = 0.10;
  if (!nh_private_.getParam ("kf_dist_angular", kf_dist_angular_))
    kf_dist_angular_ = 10.0 * (M_PI / 180.0);

  kf_dist_linear_sq_ = kf_dist_linear_ * kf_dist_linear_;

  // **** What predictions are available to speed up the ICP?
  // 1) imu - [theta] from imu yaw angle - /imu topic
  // 2) odom - [x, y, theta] from wheel odometry - /odom topic
  // 3) vel - [x, y, theta] from velocity predictor - see alpha-beta predictors - /vel topic
  // If more than one is enabled, priority is imu > odom > vel

  if (!nh_private_.getParam ("use_imu", use_imu_))
    use_imu_ = true;
  if (!nh_private_.getParam ("use_odom", use_odom_))
    use_odom_ = true;
  if (!nh_private_.getParam ("use_tf", use_tf_))
    use_tf_ = true;
  if (!nh_private_.getParam ("tf_timeout", tf_timeout_))
    tf_timeout_ = 0.1;
  if (!nh_private_.getParam ("use_vel", use_vel_))
    use_vel_ = false;

  // **** Are velocity input messages stamped?
  // if false, will subscribe to Twist msgs on /vel
  // if true, will subscribe to TwistStamped msgs on /vel
  if (!nh_private_.getParam ("stamped_vel", stamped_vel_))
    stamped_vel_ = false;

  // **** How to publish the output?
  // tf (fixed_frame->base_frame),
  // pose message (pose of base frame in the fixed frame)

  if (!nh_private_.getParam ("publish_tf", publish_tf_))
    publish_tf_ = true;
  if (!nh_private_.getParam ("publish_pose", publish_pose_))
    publish_pose_ = true;
  if (!nh_private_.getParam ("publish_pose_stamped", publish_pose_stamped_))
    publish_pose_stamped_ = false;
  if (!nh_private_.getParam ("publish_pose_with_covariance", publish_pose_with_covariance_))
    publish_pose_with_covariance_ = false;
  if (!nh_private_.getParam ("publish_pose_with_covariance_stamped", publish_pose_with_covariance_stamped_))
    publish_pose_with_covariance_stamped_ = false;

  if (!nh_private_.getParam("position_covariance", position_covariance_))
  {
    position_covariance_.resize(3);
    std::fill(position_covariance_.begin(), position_covariance_.end(), 1e-9);
  }

  if (!nh_private_.getParam("orientation_covariance", orientation_covariance_))
  {
    orientation_covariance_.resize(3);
    std::fill(orientation_covariance_.begin(), orientation_covariance_.end(), 1e-9);
  }
  // **** CSM parameters - comments copied from algos.h (by Andrea Censi)

  // Maximum angular displacement between scans
  if (!nh_private_.getParam ("max_angular_correction_deg", input_.max_angular_correction_deg))
    input_.max_angular_correction_deg = 45.0;

  // Maximum translation between scans (m)
  if (!nh_private_.getParam ("max_linear_correction", input_.max_linear_correction))
    input_.max_linear_correction = 0.50;

  // Maximum ICP cycle iterations
  if (!nh_private_.getParam ("max_iterations", input_.max_iterations))
    input_.max_iterations = 10;

  // A threshold for stopping (m)
  if (!nh_private_.getParam ("epsilon_xy", input_.epsilon_xy))
    input_.epsilon_xy = 0.000001;

  // A threshold for stopping (rad)
  if (!nh_private_.getParam ("epsilon_theta", input_.epsilon_theta))
    input_.epsilon_theta = 0.000001;

  // Maximum distance for a correspondence to be valid
  if (!nh_private_.getParam ("max_correspondence_dist", input_.max_correspondence_dist))
    input_.max_correspondence_dist = 0.3;

  // Noise in the scan (m)
  if (!nh_private_.getParam ("sigma", input_.sigma))
    input_.sigma = 0.010;

  // Use smart tricks for finding correspondences.
  if (!nh_private_.getParam ("use_corr_tricks", input_.use_corr_tricks))
    input_.use_corr_tricks = 1;

  // Restart: Restart if error is over threshold
  if (!nh_private_.getParam ("restart", input_.restart))
    input_.restart = 0;

  // Restart: Threshold for restarting
  if (!nh_private_.getParam ("restart_threshold_mean_error", input_.restart_threshold_mean_error))
    input_.restart_threshold_mean_error = 0.01;

  // Restart: displacement for restarting. (m)
  if (!nh_private_.getParam ("restart_dt", input_.restart_dt))
    input_.restart_dt = 1.0;

  // Restart: displacement for restarting. (rad)
  if (!nh_private_.getParam ("restart_dtheta", input_.restart_dtheta))
    input_.restart_dtheta = 0.1;

  // Max distance for staying in the same clustering
  if (!nh_private_.getParam ("clustering_threshold", input_.clustering_threshold))
    input_.clustering_threshold = 0.25;

  // Number of neighbour rays used to estimate the orientation
  if (!nh_private_.getParam ("orientation_neighbourhood", input_.orientation_neighbourhood))
    input_.orientation_neighbourhood = 20;

  // If 0, it's vanilla ICP
  if (!nh_private_.getParam ("use_point_to_line_distance", input_.use_point_to_line_distance))
    input_.use_point_to_line_distance = 1;

  // Discard correspondences based on the angles
  if (!nh_private_.getParam ("do_alpha_test", input_.do_alpha_test))
    input_.do_alpha_test = 0;

  // Discard correspondences based on the angles - threshold angle, in degrees
  if (!nh_private_.getParam ("do_alpha_test_thresholdDeg", input_.do_alpha_test_thresholdDeg))
    input_.do_alpha_test_thresholdDeg = 20.0;

  // Percentage of correspondences to consider: if 0.9,
  // always discard the top 10% of correspondences with more error
  if (!nh_private_.getParam ("outliers_maxPerc", input_.outliers_maxPerc))
    input_.outliers_maxPerc = 0.90;

  // Parameters describing a simple adaptive algorithm for discarding.
  //  1) Order the errors.
  //  2) Choose the percentile according to outliers_adaptive_order.
  //     (if it is 0.7, get the 70% percentile)
  //  3) Define an adaptive threshold multiplying outliers_adaptive_mult
  //     with the value of the error at the chosen percentile.
  //  4) Discard correspondences over the threshold.
  //  This is useful to be conservative; yet remove the biggest errors.
  if (!nh_private_.getParam ("outliers_adaptive_order", input_.outliers_adaptive_order))
    input_.outliers_adaptive_order = 0.7;

  if (!nh_private_.getParam ("outliers_adaptive_mult", input_.outliers_adaptive_mult))
    input_.outliers_adaptive_mult = 2.0;

  // If you already have a guess of the solution, you can compute the polar angle
  // of the points of one scan in the new position. If the polar angle is not a monotone
  // function of the readings index, it means that the surface is not visible in the
  // next position. If it is not visible, then we don't use it for matching.
  if (!nh_private_.getParam ("do_visibility_test", input_.do_visibility_test))
    input_.do_visibility_test = 0;

  // no two points in laser_sens can have the same corr.
  if (!nh_private_.getParam ("outliers_remove_doubles", input_.outliers_remove_doubles))
    input_.outliers_remove_doubles = 1;

  // If 1, computes the covariance of ICP using the method http://purl.org/censi/2006/icpcov
  if (!nh_private_.getParam ("do_compute_covariance", input_.do_compute_covariance))
    input_.do_compute_covariance = 0;

  // Checks that find_correspondences_tricks gives the right answer
  if (!nh_private_.getParam ("debug_verify_tricks", input_.debug_verify_tricks))
    input_.debug_verify_tricks = 0;

  // If 1, the field 'true_alpha' (or 'alpha') in the first scan is used to compute the
  // incidence beta, and the factor (1/cos^2(beta)) used to weight the correspondence.");
  if (!nh_private_.getParam ("use_ml_weights", input_.use_ml_weights))
    input_.use_ml_weights = 0;

  // If 1, the field 'readings_sigma' in the second scan is used to weight the
  // correspondence by 1/sigma^2
  if (!nh_private_.getParam ("use_sigma_weights", input_.use_sigma_weights))
    input_.use_sigma_weights = 0;
}

void LaserScanMatcher::imuCallback(const sensor_msgs::Imu::ConstPtr& imu_msg)
{
  boost::mutex::scoped_lock(mutex_);
  latest_imu_msg_ = *imu_msg;
  if (!received_imu_)
  {
    tf::quaternionMsgToTF(imu_msg->orientation, last_used_imu_orientation_);
    received_imu_ = true;
  }
}

void LaserScanMatcher::odomCallback(const nav_msgs::Odometry::ConstPtr& odom_msg)
{
  boost::mutex::scoped_lock(mutex_);
  latest_odom_msg_ = *odom_msg;
  if (!received_odom_)
  {
    tf::poseMsgToTF(odom_msg->pose.pose, last_used_odom_pose_);
    received_odom_ = true;
  }
}

void LaserScanMatcher::velCallback(const geometry_msgs::Twist::ConstPtr& twist_msg)
{
  boost::mutex::scoped_lock(mutex_);
  latest_vel_msg_ = *twist_msg;

  received_vel_ = true;
}

void LaserScanMatcher::velStmpCallback(const geometry_msgs::TwistStamped::ConstPtr& twist_msg)
{
  boost::mutex::scoped_lock(mutex_);
  latest_vel_msg_ = twist_msg->twist;

  received_vel_ = true;
}

void LaserScanMatcher::cloudCallback (const PointCloudT::ConstPtr& cloud)
{
  // **** if first scan, cache the tf from base to the scanner

  std_msgs::Header cloud_header = pcl_conversions::fromPCL(cloud->header);

  if (!initialized_)
  {
    // cache the static tf from base to laser
    if (!getBaseLaserTransform(cloud_header.frame_id))
    {
      ROS_WARN("Skipping scan");
      return;
    }

    PointCloudToLDP(cloud, prev_ldp_scan_);
    last_icp_time_ = cloud_header.stamp;
    initialized_ = true;
  }

  LDP curr_ldp_scan;
  PointCloudToLDP(cloud, curr_ldp_scan);
  processScan(curr_ldp_scan, cloud_header.stamp);
}

void LaserScanMatcher::scanCallback (const sensor_msgs::LaserScan::ConstPtr& scan_msg)
{
  // **** if first scan, cache the tf from base to the scanner

  if (!initialized_)
  {
    createCache(scan_msg);    // caches the sin and cos of all angles

    // cache the static transform between the base and laser
    if (!getBaseLaserTransform(scan_msg->header.frame_id))
    {
      ROS_WARN("Skipping scan");
      return;
    }

    laserScanToLDP(scan_msg, prev_ldp_scan_);
    last_icp_time_ = scan_msg->header.stamp;
    initialized_ = true;
  }

  LDP curr_ldp_scan;
  laserScanToLDP(scan_msg, curr_ldp_scan);
  processScan(curr_ldp_scan, scan_msg->header.stamp);
}

void LaserScanMatcher::processScan(LDP& curr_ldp_scan, const ros::Time& time)
{
  ros::WallTime start = ros::WallTime::now();

  // CSM is used in the following way:
  // The scans are always in the laser frame
  // The reference scan (prevLDPcan_) has a pose of [0, 0, 0]
  // The new scan (currLDPScan) has a pose equal to the movement
  // of the laser in the laser frame since the last scan
  // The computed correction is then propagated using the tf machinery

  prev_ldp_scan_->odometry[0] = 0.0;
  prev_ldp_scan_->odometry[1] = 0.0;
  prev_ldp_scan_->odometry[2] = 0.0;

  prev_ldp_scan_->estimate[0] = 0.0;
  prev_ldp_scan_->estimate[1] = 0.0;
  prev_ldp_scan_->estimate[2] = 0.0;

  prev_ldp_scan_->true_pose[0] = 0.0;
  prev_ldp_scan_->true_pose[1] = 0.0;
  prev_ldp_scan_->true_pose[2] = 0.0;

  input_.laser_ref  = prev_ldp_scan_;
  input_.laser_sens = curr_ldp_scan;

  // **** estimated change since last scan

  // get the predicted offset of the scan base pose from the last scan base pose
  tf::Transform pred_last_base_offset = getPrediction(time);

  // calculate the predicted scan base pose by applying the predicted offset to the last scan base pose
  tf::Transform pred_base_in_fixed = last_base_in_fixed_ * pred_last_base_offset;

  // calculate the offset between the keyframe base pose and predicted scan base pose
  tf::Transform pred_keyframe_base_offset = keyframe_base_in_fixed_.inverse() * pred_base_in_fixed;

  // convert the predicted offset from the keyframe base frame to be in the keyframe laser frame
  tf::Transform pred_keyframe_laser_offset = laser_from_base_ * pred_keyframe_base_offset * base_from_laser_ ;

  input_.first_guess[0] = pred_keyframe_laser_offset.getOrigin().getX();
  input_.first_guess[1] = pred_keyframe_laser_offset.getOrigin().getY();
  input_.first_guess[2] = tf::getYaw(pred_keyframe_laser_offset.getRotation());

  // If they are non-Null, free covariance gsl matrices to avoid leaking memory
  if (output_.cov_x_m)
  {
    gsl_matrix_free(output_.cov_x_m);
    output_.cov_x_m = 0;
  }
  if (output_.dx_dy1_m)
  {
    gsl_matrix_free(output_.dx_dy1_m);
    output_.dx_dy1_m = 0;
  }
  if (output_.dx_dy2_m)
  {
    gsl_matrix_free(output_.dx_dy2_m);
    output_.dx_dy2_m = 0;
  }

  // *** scan match - using point to line icp from CSM

  sm_icp(&input_, &output_);
  tf::Transform meas_keyframe_base_offset;

  if (output_.valid)
  {

    // the measured offset of the scan from the keyframe in the keyframe laser frame
    tf::Transform meas_keyframe_laser_offset;
    createTfFromXYTheta(output_.x[0], output_.x[1], output_.x[2], meas_keyframe_laser_offset);

    // convert the measured offset from the keyframe laser frame to the keyframe base frame
    meas_keyframe_base_offset = base_from_laser_ * meas_keyframe_laser_offset * laser_from_base_;

    // calculate the measured pose of the scan in the fixed frame
    last_base_in_fixed_ = keyframe_base_in_fixed_ * meas_keyframe_base_offset;

    // **** publish

    Eigen::Matrix2f xy_cov = Eigen::Matrix2f::Zero();
    float yaw_cov = 0.0;
    if (input_.do_compute_covariance)
    {
      // get covariance from ICP
      xy_cov(0, 0) = gsl_matrix_get(output_.cov_x_m, 0, 0);
      xy_cov(0, 1) = gsl_matrix_get(output_.cov_x_m, 0, 1);
      xy_cov(1, 0) = gsl_matrix_get(output_.cov_x_m, 1, 0);
      xy_cov(1, 1) = gsl_matrix_get(output_.cov_x_m, 1, 1);
      yaw_cov = gsl_matrix_get(output_.cov_x_m, 2, 2);

      // rotate xy covariance from the keyframe into odom frame
      auto rotation = getLaserRotation(keyframe_base_in_fixed_);
      xy_cov = rotation * xy_cov * rotation.transpose();
    }
    else {
      xy_cov(0, 0) = position_covariance_[0];
      xy_cov(1, 1) = position_covariance_[1];
      yaw_cov = orientation_covariance_[2];
    }

    if (publish_pose_)
    {
      // unstamped Pose2D message
      geometry_msgs::Pose2D::Ptr pose_msg;
      pose_msg = boost::make_shared<geometry_msgs::Pose2D>();
      pose_msg->x = last_base_in_fixed_.getOrigin().getX();
      pose_msg->y = last_base_in_fixed_.getOrigin().getY();
      pose_msg->theta = tf::getYaw(last_base_in_fixed_.getRotation());
      pose_publisher_.publish(pose_msg);
    }
    if (publish_pose_stamped_)
    {
      // stamped Pose message
      geometry_msgs::PoseStamped::Ptr pose_stamped_msg;
      pose_stamped_msg = boost::make_shared<geometry_msgs::PoseStamped>();

      pose_stamped_msg->header.stamp    = time;
      pose_stamped_msg->header.frame_id = fixed_frame_;

      tf::poseTFToMsg(last_base_in_fixed_, pose_stamped_msg->pose);

      pose_stamped_publisher_.publish(pose_stamped_msg);
    }
    if (publish_pose_with_covariance_)
    {
      // unstamped PoseWithCovariance message
      geometry_msgs::PoseWithCovariance::Ptr pose_with_covariance_msg;
      pose_with_covariance_msg = boost::make_shared<geometry_msgs::PoseWithCovariance>();
      tf::poseTFToMsg(last_base_in_fixed_, pose_with_covariance_msg->pose);

      pose_with_covariance_msg->covariance = boost::assign::list_of
        (xy_cov(0, 0)) (xy_cov(0, 1)) (0) (0) (0) (0)
        (xy_cov(1, 0)) (xy_cov(1, 1)) (0) (0) (0) (0)
        (0) (0) (0) (0) (0) (0)
        (0) (0) (0) (0) (0) (0)
        (0) (0) (0) (0) (0) (0)
        (0) (0) (0) (0) (0) (yaw_cov);

      pose_with_covariance_publisher_.publish(pose_with_covariance_msg);
    }
    if (publish_pose_with_covariance_stamped_)
    {
      // stamped Pose message
      geometry_msgs::PoseWithCovarianceStamped::Ptr pose_with_covariance_stamped_msg;
      pose_with_covariance_stamped_msg = boost::make_shared<geometry_msgs::PoseWithCovarianceStamped>();

      pose_with_covariance_stamped_msg->header.stamp    = time;
      pose_with_covariance_stamped_msg->header.frame_id = fixed_frame_;

      tf::poseTFToMsg(last_base_in_fixed_, pose_with_covariance_stamped_msg->pose.pose);

      pose_with_covariance_stamped_msg->pose.covariance = boost::assign::list_of
        (xy_cov(0, 0)) (xy_cov(0, 1)) (0) (0) (0) (0)
        (xy_cov(1, 0)) (xy_cov(1, 1)) (0) (0) (0) (0)
        (0) (0) (0) (0) (0) (0)
        (0) (0) (0) (0) (0) (0)
        (0) (0) (0) (0) (0) (0)
        (0) (0) (0) (0) (0) (yaw_cov);

      pose_with_covariance_stamped_publisher_.publish(pose_with_covariance_stamped_msg);
    }

    if (publish_tf_)
    {
      tf::StampedTransform transform_msg (last_base_in_fixed_, time, fixed_frame_, base_frame_);
      tf_broadcaster_.sendTransform (transform_msg);
    }
  }
  else
  {
    meas_keyframe_base_offset.setIdentity();
    ROS_WARN("Error in scan matching");
  }

  // **** swap old and new

  if (newKeyframeNeeded(meas_keyframe_base_offset))
  {
    // generate a keyframe
    ld_free(prev_ldp_scan_);
    prev_ldp_scan_ = curr_ldp_scan;
    keyframe_base_in_fixed_ = last_base_in_fixed_;
  }
  else
  {
    ld_free(curr_ldp_scan);
  }

  last_icp_time_ = time;

  // **** statistics

  double dur = (ros::WallTime::now() - start).toSec() * 1e3;
  ROS_DEBUG("Scan matcher total duration: %.1f ms", dur);
}

bool LaserScanMatcher::newKeyframeNeeded(const tf::Transform& d)
{
  if (fabs(tf::getYaw(d.getRotation())) > kf_dist_angular_) return true;

  double x = d.getOrigin().getX();
  double y = d.getOrigin().getY();
  if (x*x + y*y > kf_dist_linear_sq_) return true;

  return false;
}

void LaserScanMatcher::PointCloudToLDP(const PointCloudT::ConstPtr& cloud,
                                             LDP& ldp)
{
  double max_d2 = cloud_res_ * cloud_res_;

  PointCloudT cloud_f;

  cloud_f.points.push_back(cloud->points[0]);

  for (unsigned int i = 1; i < cloud->points.size(); ++i)
  {
    const PointT& pa = cloud_f.points[cloud_f.points.size() - 1];
    const PointT& pb = cloud->points[i];

    double dx = pa.x - pb.x;
    double dy = pa.y - pb.y;
    double d2 = dx*dx + dy*dy;

    if (d2 > max_d2)
    {
      cloud_f.points.push_back(pb);
    }
  }

  unsigned int n = cloud_f.points.size();

  ldp = ld_alloc_new(n);

  for (unsigned int i = 0; i < n; i++)
  {
    // calculate position in laser frame
    if (is_nan(cloud_f.points[i].x) || is_nan(cloud_f.points[i].y))
    {
      ROS_WARN("Laser Scan Matcher: Cloud input contains NaN values. \
                Please use a filtered cloud input.");
    }
    else
    {
      double r = sqrt(cloud_f.points[i].x * cloud_f.points[i].x +
                      cloud_f.points[i].y * cloud_f.points[i].y);

      if (r > cloud_range_min_ && r < cloud_range_max_)
      {
        ldp->valid[i] = 1;
        ldp->readings[i] = r;
      }
      else
      {
        ldp->valid[i] = 0;
        ldp->readings[i] = -1;  // for invalid range
      }
    }

    ldp->theta[i] = atan2(cloud_f.points[i].y, cloud_f.points[i].x);
    ldp->cluster[i]  = -1;
  }

  ldp->min_theta = ldp->theta[0];
  ldp->max_theta = ldp->theta[n-1];

  ldp->odometry[0] = 0.0;
  ldp->odometry[1] = 0.0;
  ldp->odometry[2] = 0.0;

  ldp->true_pose[0] = 0.0;
  ldp->true_pose[1] = 0.0;
  ldp->true_pose[2] = 0.0;
}

void LaserScanMatcher::laserScanToLDP(const sensor_msgs::LaserScan::ConstPtr& scan_msg,
                                            LDP& ldp)
{
  unsigned int n = scan_msg->ranges.size();
  ldp = ld_alloc_new(n);

  for (unsigned int i = 0; i < n; i++)
  {
    // calculate position in laser frame

    double r = scan_msg->ranges[i];

    if (r > scan_msg->range_min && r < scan_msg->range_max)
    {
      // fill in laser scan data

      ldp->valid[i] = 1;
      ldp->readings[i] = r;
    }
    else
    {
      ldp->valid[i] = 0;
      ldp->readings[i] = -1;  // for invalid range
    }

    ldp->theta[i]    = scan_msg->angle_min + i * scan_msg->angle_increment;

    ldp->cluster[i]  = -1;
  }

  ldp->min_theta = ldp->theta[0];
  ldp->max_theta = ldp->theta[n-1];

  ldp->odometry[0] = 0.0;
  ldp->odometry[1] = 0.0;
  ldp->odometry[2] = 0.0;

  ldp->true_pose[0] = 0.0;
  ldp->true_pose[1] = 0.0;
  ldp->true_pose[2] = 0.0;
}

void LaserScanMatcher::createCache (const sensor_msgs::LaserScan::ConstPtr& scan_msg)
{
  a_cos_.clear();
  a_sin_.clear();

  for (unsigned int i = 0; i < scan_msg->ranges.size(); ++i)
  {
    double angle = scan_msg->angle_min + i * scan_msg->angle_increment;
    a_cos_.push_back(cos(angle));
    a_sin_.push_back(sin(angle));
  }

  input_.min_reading = scan_msg->range_min;
  input_.max_reading = scan_msg->range_max;
}

bool LaserScanMatcher::getBaseLaserTransform(const std::string& frame_id)
{
  tf::StampedTransform base_from_laser;
  try
  {
    tf_listener_.waitForTransform(base_frame_, frame_id, ros::Time(0), ros::Duration(tf_timeout_));
    tf_listener_.lookupTransform (base_frame_, frame_id, ros::Time(0), base_from_laser);
  }
  catch (const tf::TransformException& ex)
  {
    ROS_WARN("Could not get initial transform from base to laser frame, %s", ex.what());
    return false;
  }
  base_from_laser_ = base_from_laser;
  laser_from_base_ = base_from_laser_.inverse();

  return true;
}

// returns the predicted offset from the base pose of the last scan
tf::Transform LaserScanMatcher::getPrediction(const ros::Time& stamp)
{
  boost::mutex::scoped_lock(mutex_);

  // **** base case - no input available, use zero-motion model
  tf::Transform pred_last_base_offset = tf::Transform::getIdentity();

  // **** use velocity (for example from ab-filter)
  if (use_vel_)
  {
    double dt = (stamp - last_icp_time_).toSec();
    // NOTE: this assumes the velocity is in the base frame and that the base
    //       and laser frames share the same x,y and z axes
    double pr_ch_x = dt * latest_vel_msg_.linear.x;
    double pr_ch_y = dt * latest_vel_msg_.linear.y;
    double pr_ch_a = dt * latest_vel_msg_.angular.z;

    createTfFromXYTheta(pr_ch_x, pr_ch_y, pr_ch_a, pred_last_base_offset);
  }

  // **** use wheel odometry
  if (use_odom_ && received_odom_)
  {
    // NOTE: this assumes the odometry is in the base frame
    tf::Transform latest_odom_pose;
    tf::poseMsgToTF(latest_odom_msg_.pose.pose, latest_odom_pose);

    pred_last_base_offset = last_used_odom_pose_.inverse() * latest_odom_pose;

    last_used_odom_pose_ = latest_odom_pose;
  }

  // **** use imu
  if (use_imu_ && received_imu_)
  {
    // NOTE: this assumes the imu is in the base frame
    tf::Quaternion latest_imu_orientation;
    tf::quaternionMsgToTF(latest_imu_msg_.orientation, latest_imu_orientation);

    tf::Quaternion imu_orientation_offset = last_used_imu_orientation_.inverse() * latest_imu_orientation;
    pred_last_base_offset.setRotation(imu_orientation_offset);

    last_used_imu_orientation_ = latest_imu_orientation;
  }

  // **** use tf
  if (use_tf_)
  {
    tf::StampedTransform pred_last_base_offset_tf;
    try
    {
      tf_listener_.waitForTransform(base_frame_, last_icp_time_, base_frame_, stamp, fixed_frame_, ros::Duration(tf_timeout_));
      tf_listener_.lookupTransform (base_frame_, last_icp_time_, base_frame_, stamp, fixed_frame_, pred_last_base_offset_tf);
      pred_last_base_offset = pred_last_base_offset_tf;
    }
    catch (const tf::TransformException& ex)
    {
      ROS_WARN("Could not get base to fixed frame transform, %s", ex.what());
    }
  }

  return pred_last_base_offset;
}

void LaserScanMatcher::createTfFromXYTheta(
  double x, double y, double theta, tf::Transform& t)
{
  t.setOrigin(tf::Vector3(x, y, 0.0));
  tf::Quaternion q;
  q.setRPY(0.0, 0.0, theta);
  t.setRotation(q);
}

Eigen::Matrix2f LaserScanMatcher::getLaserRotation(const tf::Transform& odom_pose) const {
  tf::Transform laser_in_fixed = odom_pose * laser_from_base_;
  tf::Matrix3x3 fixed_from_laser_rot(laser_in_fixed.getRotation());
  double r,p,y;
  fixed_from_laser_rot.getRPY(r, p, y);
  Eigen::Rotation2Df t(y);
  return t.toRotationMatrix();
}

} // namespace scan_tools