i/rtabmap_ros
1
0
Fork 0
mirror of https://github.com/introlab/rtabmap_ros.git synced 2026-09-12 22:30:19 +08:00
Code Issues Packages Projects Releases Wiki Activity

Refactored obstacles_detection nodelet to use OccupancyGrid segmentation approach (so using exactly the same parameter names as rtabmap node for the map)

Browse Source
This commit is contained in:
matlabbe
2016-08-31 12:45:55 -04:00
parent da2b87bb28
commit 750fd61daa
7 changed files with 562 additions and 452 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CMakeLists.txt
+1 -1
View File
@@ -160,7 +160,7 @@ SET(rtabmap_ros_lib_src
src/nodelets/point_cloud_xyz.cpp
src/nodelets/disparity_to_depth.cpp
src/nodelets/obstacles_detection.cpp
src/nodelets/obstacles_detection_indoor.cpp
src/nodelets/obstacles_detection_old.cpp
src/nodelets/point_cloud_aggregator.cpp
src/OdometryROS.cpp
src/MsgConversion.cpp
include/rtabmap_ros/CoreWrapper.h
-1
View File
@@ -283,7 +283,6 @@ private:
double genScanMinDepth_;
int scanCloudMaxPoints_;
int scanCloudNormalK_;
bool flipScan_;
rtabmap::Transform mapToOdom_;
boost::mutex mapToOdomMutex_;
launch/tests/test_obstacles_detection.launch
-3
View File
@@ -1,8 +1,6 @@
<launch>
<!-- Use stereo_outdoorA.bag for testing -->
<arg name="optimize_for_close_objects" default="false" />
<include file="$(find rtabmap_ros)/launch/demo/demo_stereo_outdoor.launch"/>
<group ns="/stereo_camera" >
@@ -23,7 +21,6 @@
<param name="wait_for_transform" type="bool" value="true"/>
<param name="min_cluster_size" type="int" value="20"/>
<param name="max_obstacles_height" type="double" value="0.0"/>
<param name="optimize_for_close_objects" type="bool" value="$(arg optimize_for_close_objects)"/>
</node>
</group>
nodelet_plugins.xml
+2 -2
View File
@@ -88,8 +88,8 @@
</description>
</class>
<class name="rtabmap_ros/obstacles_detection_indoor"
type="rtabmap_ros::ObstaclesDetectionIndoor"
<class name="rtabmap_ros/obstacles_detection_old"
type="rtabmap_ros::ObstaclesDetectionOld"
base_class_type="nodelet::Nodelet">
<description>
This is my nodelet.
src/nodelets/obstacles_detection.cpp
+202 -178
View File
@@ -36,28 +36,11 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#include <tf/transform_listener.h>
#include <sensor_msgs/PointCloud2.h>
#include <sensor_msgs/Image.h>
#include <sensor_msgs/image_encodings.h>
#include <sensor_msgs/CameraInfo.h>
#include <stereo_msgs/DisparityImage.h>
#include <image_transport/image_transport.h>
#include <image_transport/subscriber_filter.h>
#include <image_geometry/pinhole_camera_model.h>
#include <message_filters/sync_policies/approximate_time.h>
#include <message_filters/subscriber.h>
#include <cv_bridge/cv_bridge.h>
#include <opencv2/highgui/highgui.hpp>
#include <rtabmap_ros/MsgConversion.h>
#include "rtabmap/core/util3d.h"
#include "rtabmap/core/util3d_filtering.h"
#include "rtabmap/core/util3d_mapping.h"
#include "rtabmap/core/util3d_transforms.h"
#include "rtabmap/core/OccupancyGrid.h"
#include "rtabmap/utilite/UStl.h"
namespace rtabmap_ros
{
@@ -67,51 +50,154 @@ class ObstaclesDetection : public nodelet::Nodelet
public:
ObstaclesDetection() :
frameId_("base_link"),
normalKSearch_(20),
groundNormalAngle_(M_PI_4),
clusterRadius_(0.05),
minClusterSize_(20),
maxObstaclesHeight_(0.0), // if<=0.0 -> disabled
maxGroundHeight_(0.0), // if<=0.0 -> disabled, used only if detect_flat_obstacles is true
segmentFlatObstacles_(false),
waitForTransform_(false),
optimizeForCloseObjects_(false),
projVoxelSize_(0.01)
waitForTransform_(false)
{}
virtual ~ObstaclesDetection()
{}
private:
void parameterMoved(
ros::NodeHandle & nh,
const std::string & rosName,
const std::string & parameterName,
rtabmap::ParametersMap & parameters)
{
if(nh.hasParam(rosName))
{
rtabmap::ParametersMap gridParameters = rtabmap::Parameters::getDefaultParameters("Grid");
rtabmap::ParametersMap::const_iterator iter =gridParameters.find(parameterName);
if(iter != gridParameters.end())
{
NODELET_ERROR("obstacles_detection: Parameter \"%s\" has moved from "
"rtabmap_ros to rtabmap library. Use "
"parameter \"%s\" instead. The value is still "
"copied to new parameter name.",
rosName.c_str(),
parameterName.c_str());
std::string type = rtabmap::Parameters::getType(parameterName);
if(type.compare("float") || type.compare("double"))
{
double v = uStr2Double(iter->second);
nh.getParam(rosName, v);
parameters.insert(rtabmap::ParametersPair(parameterName, uNumber2Str(v)));
}
else if(type.compare("int") || type.compare("unsigned int"))
{
int v = uStr2Int(iter->second);
nh.getParam(rosName, v);
parameters.insert(rtabmap::ParametersPair(parameterName, uNumber2Str(v)));
}
else
{
NODELET_ERROR("Not handled type \"%s\" for parameter \"%s\"", type.c_str(), parameterName.c_str());
}
}
else
{
NODELET_ERROR("Parameter \"%s\" not found in default parameters.", parameterName.c_str());
}
}
}
virtual void onInit()
{
ROS_DEBUG("_"); // not sure why, but all NODELET_*** log are not shown if a normal ROS_*** is not called!?
ros::NodeHandle & nh = getNodeHandle();
ros::NodeHandle & pnh = getPrivateNodeHandle();
int queueSize = 10;
pnh.param("queue_size", queueSize, queueSize);
pnh.param("frame_id", frameId_, frameId_);
pnh.param("normal_k", normalKSearch_, normalKSearch_);
pnh.param("ground_normal_angle", groundNormalAngle_, groundNormalAngle_);
if(pnh.hasParam("normal_estimation_radius") && !pnh.hasParam("cluster_radius"))
{
NODELET_WARN("Parameter \"normal_estimation_radius\" has been renamed "
"to \"cluster_radius\"! Your value is still copied to "
"corresponding parameter. Instead of normal radius, nearest neighbors count "
"\"normal_k\" is used instead (default 20).");
pnh.param("normal_estimation_radius", clusterRadius_, clusterRadius_);
}
else
{
pnh.param("cluster_radius", clusterRadius_, clusterRadius_);
}
pnh.param("min_cluster_size", minClusterSize_, minClusterSize_);
pnh.param("max_obstacles_height", maxObstaclesHeight_, maxObstaclesHeight_);
pnh.param("max_ground_height", maxGroundHeight_, maxGroundHeight_);
pnh.param("detect_flat_obstacles", segmentFlatObstacles_, segmentFlatObstacles_);
pnh.param("map_frame_id", mapFrameId_, mapFrameId_);
pnh.param("wait_for_transform", waitForTransform_, waitForTransform_);
pnh.param("optimize_for_close_objects", optimizeForCloseObjects_, optimizeForCloseObjects_);
pnh.param("proj_voxel_size", projVoxelSize_, projVoxelSize_);
if(pnh.hasParam("optimize_for_close_objects"))
{
NODELET_ERROR("\"optimize_for_close_objects\" parameter doesn't exist "
"anymore. Use rtabmap_ros/obstacles_detection_old nodelet to use "
"the old interface.");
}
rtabmap::ParametersMap parameters;
// Backward compatibility
for(std::map<std::string, std::pair<bool, std::string> >::const_iterator iter=rtabmap::Parameters::getRemovedParameters().begin();
iter!=rtabmap::Parameters::getRemovedParameters().end();
++iter)
{
std::string vStr;
if(pnh.getParam(iter->first, vStr))
{
if(iter->second.first)
{
// can be migrated
uInsert(parameters, rtabmap::ParametersPair(iter->second.second, vStr));
NODELET_ERROR("obstacles_detection: Parameter name changed: \"%s\" -> \"%s\". Please update your launch file accordingly. Value \"%s\" is still set to the new parameter name.",
iter->first.c_str(), iter->second.second.c_str(), vStr.c_str());
}
else
{
if(iter->second.second.empty())
{
NODELET_ERROR("obstacles_detection: Parameter \"%s\" doesn't exist anymore!",
iter->first.c_str());
}
else
{
NODELET_ERROR("obstacles_detection: Parameter \"%s\" doesn't exist anymore! You may look at this similar parameter: \"%s\"",
iter->first.c_str(), iter->second.second.c_str());
}
}
}
}
rtabmap::ParametersMap gridParameters2 = rtabmap::Parameters::getDefaultParameters();
rtabmap::ParametersMap gridParameters = rtabmap::Parameters::getDefaultParameters("Grid");
for(rtabmap::ParametersMap::iterator iter=gridParameters.begin(); iter!=gridParameters.end(); ++iter)
{
std::string vStr;
bool vBool;
int vInt;
double vDouble;
if(pnh.getParam(iter->first, vStr))
{
NODELET_INFO("obstacles_detection: Setting parameter \"%s\"=\"%s\"", iter->first.c_str(), vStr.c_str());
iter->second = vStr;
}
else if(pnh.getParam(iter->first, vBool))
{
NODELET_INFO("obstacles_detection: Setting parameter \"%s\"=\"%s\"", iter->first.c_str(), uBool2Str(vBool).c_str());
iter->second = uBool2Str(vBool);
}
else if(pnh.getParam(iter->first, vDouble))
{
NODELET_INFO("obstacles_detection: Setting parameter \"%s\"=\"%s\"", iter->first.c_str(), uNumber2Str(vDouble).c_str());
iter->second = uNumber2Str(vDouble);
}
else if(pnh.getParam(iter->first, vInt))
{
NODELET_INFO("obstacles_detection: Setting parameter \"%s\"=\"%s\"", iter->first.c_str(), uNumber2Str(vInt).c_str());
iter->second = uNumber2Str(vInt);
}
}
uInsert(parameters, gridParameters);
parameterMoved(pnh, "proj_voxel_size", rtabmap::Parameters::kGridCellSize(), parameters);
parameterMoved(pnh, "ground_normal_angle", rtabmap::Parameters::kGridMaxGroundAngle(), parameters);
parameterMoved(pnh, "min_cluster_size", rtabmap::Parameters::kGridMinClusterSize(), parameters);
parameterMoved(pnh, "normal_estimation_radius", rtabmap::Parameters::kGridClusterRadius(), parameters);
parameterMoved(pnh, "cluster_radius", rtabmap::Parameters::kGridClusterRadius(), parameters);
parameterMoved(pnh, "max_obstacles_height", rtabmap::Parameters::kGridMaxObstacleHeight(), parameters);
parameterMoved(pnh, "max_ground_height", rtabmap::Parameters::kGridMaxGroundHeight(), parameters);
parameterMoved(pnh, "detect_flat_obstacles", rtabmap::Parameters::kGridFlatObstacleDetected(), parameters);
parameterMoved(pnh, "normal_k", rtabmap::Parameters::kGridNormalK(), parameters);
UASSERT(uContains(parameters, rtabmap::Parameters::kGridMapFrameProjection()));
if(uStr2Bool(parameters.at(rtabmap::Parameters::kGridMapFrameProjection())) && mapFrameId_.empty())
{
NODELET_ERROR("obstacles_detection: Parameter \"%s\" is true but map_frame_id is not set!", rtabmap::Parameters::kGridMapFrameProjection().c_str());
}
cloudSub_ = nh.subscribe("cloud", 1, &ObstaclesDetection::callback, this);
@@ -153,8 +239,32 @@ private:
return;
}
pcl::PointCloud<pcl::PointXYZ>::Ptr originalCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::fromROSMsg(*cloudMsg, *originalCloud);
rtabmap::Transform pose = rtabmap::Transform::getIdentity();
if(!mapFrameId_.empty())
{
try
{
if(waitForTransform_)
{
if(!tfListener_.waitForTransform(mapFrameId_, frameId_, cloudMsg->header.stamp, ros::Duration(1)))
{
NODELET_ERROR("Could not get transform from %s to %s after 1 second!", mapFrameId_.c_str(), frameId_.c_str());
return;
}
}
tf::StampedTransform tmp;
tfListener_.lookupTransform(mapFrameId_, frameId_, cloudMsg->header.stamp, tmp);
pose = rtabmap_ros::transformFromTF(tmp);
}
catch(tf::TransformException & ex)
{
NODELET_ERROR("%s",ex.what());
return;
}
}
pcl::PointCloud<pcl::PointXYZ>::Ptr inputCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::fromROSMsg(*cloudMsg, *inputCloud);
//Common variables for all strategies
pcl::IndicesPtr ground, obstacles;
@@ -162,136 +272,56 @@ private:
pcl::PointCloud<pcl::PointXYZ>::Ptr groundCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::PointCloud<pcl::PointXYZ>::Ptr obstaclesCloudWithoutFlatSurfaces(new pcl::PointCloud<pcl::PointXYZ>);
if(originalCloud->size())
if(inputCloud->size())
{
originalCloud = rtabmap::util3d::transformPointCloud(originalCloud, localTransform);
if(maxObstaclesHeight_ > 0)
{
// std::numeric_limits<float>::lowest() exists only for c++11
originalCloud = rtabmap::util3d::passThrough(originalCloud, "z", std::numeric_limits<int>::min(), maxObstaclesHeight_);
}
inputCloud = rtabmap::util3d::transformPointCloud(inputCloud, localTransform);
if(originalCloud->size())
pcl::IndicesPtr flatObstacles(new std::vector<int>);
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud = grid_.segmentCloud<pcl::PointXYZ>(
inputCloud,
pcl::IndicesPtr(new std::vector<int>),
pose,
cv::Point3f(localTransform.x(), localTransform.y(), localTransform.z()),
ground,
obstacles,
&flatObstacles);
if(cloud->size() && (ground->size() || obstacles->size()))
{
if(!optimizeForCloseObjects_)
if(groundPub_.getNumSubscribers() &&
ground.get() && ground->size())
{
// This is the default strategy
pcl::IndicesPtr flatObstacles(new std::vector<int>);
rtabmap::util3d::segmentObstaclesFromGround<pcl::PointXYZ>(
originalCloud,
ground,
obstacles,
normalKSearch_,
groundNormalAngle_,
clusterRadius_,
minClusterSize_,
segmentFlatObstacles_,
maxGroundHeight_,
&flatObstacles);
if(groundPub_.getNumSubscribers() &&
ground.get() && ground->size())
{
pcl::copyPointCloud(*originalCloud, *ground, *groundCloud);
}
if((obstaclesPub_.getNumSubscribers() || projObstaclesPub_.getNumSubscribers()) &&
obstacles.get() && obstacles->size())
{
// remove flat obstacles from obstacles
std::set<int> flatObstaclesSet;
if(projObstaclesPub_.getNumSubscribers())
{
flatObstaclesSet.insert(flatObstacles->begin(), flatObstacles->end());
}
obstaclesCloud->resize(obstacles->size());
obstaclesCloudWithoutFlatSurfaces->resize(obstacles->size());
int oi=0;
for(unsigned int i=0; i<obstacles->size(); ++i)
{
obstaclesCloud->points[i] = originalCloud->at(obstacles->at(i));
if(flatObstaclesSet.size() == 0 ||
flatObstaclesSet.find(obstacles->at(i))==flatObstaclesSet.end())
{
obstaclesCloudWithoutFlatSurfaces->points[oi] = obstaclesCloud->points[i];
obstaclesCloudWithoutFlatSurfaces->points[oi].z = 0;
++oi;
}
}
obstaclesCloudWithoutFlatSurfaces->resize(oi);
if(obstaclesCloudWithoutFlatSurfaces->size() && projVoxelSize_ > 0.0)
{
obstaclesCloudWithoutFlatSurfaces = rtabmap::util3d::voxelize(obstaclesCloudWithoutFlatSurfaces, projVoxelSize_);
}
}
pcl::copyPointCloud(*cloud, *ground, *groundCloud);
}
else
if((obstaclesPub_.getNumSubscribers() || projObstaclesPub_.getNumSubscribers()) &&
obstacles.get() && obstacles->size())
{
// in this case optimizeForCloseObject_ is true:
// we divide the floor point cloud into two subsections, one for all potential floor points up to 1m
// one for potential floor points further away than 1m.
// For the points at closer range, we use a smaller normal estimation radius and ground normal angle,
// which allows to detect smaller objects, without increasing the number of false positive.
// For all other points, we use a bigger normal estimation radius (* 3.) and tolerance for the
// grond normal angle (* 2.).
pcl::PointCloud<pcl::PointXYZ>::Ptr originalCloud_near = rtabmap::util3d::passThrough(originalCloud, "x", std::numeric_limits<int>::min(), 1.);
pcl::PointCloud<pcl::PointXYZ>::Ptr originalCloud_far = rtabmap::util3d::passThrough(originalCloud, "x", 1., std::numeric_limits<int>::max());
// Part 1: segment floor and obstacles near the robot
rtabmap::util3d::segmentObstaclesFromGround<pcl::PointXYZ>(
originalCloud_near,
ground,
obstacles,
normalKSearch_,
groundNormalAngle_,
clusterRadius_,
minClusterSize_,
segmentFlatObstacles_,
maxGroundHeight_);
if(groundPub_.getNumSubscribers() && ground.get() && ground->size())
// remove flat obstacles from obstacles
std::set<int> flatObstaclesSet;
if(projObstaclesPub_.getNumSubscribers())
{
pcl::copyPointCloud(*originalCloud_near, *ground, *groundCloud);
ground->clear();
flatObstaclesSet.insert(flatObstacles->begin(), flatObstacles->end());
}
if((obstaclesPub_.getNumSubscribers() || projObstaclesPub_.getNumSubscribers()) && obstacles.get() && obstacles->size())
obstaclesCloud->resize(obstacles->size());
obstaclesCloudWithoutFlatSurfaces->resize(obstacles->size());
int oi=0;
for(unsigned int i=0; i<obstacles->size(); ++i)
{
pcl::copyPointCloud(*originalCloud_near, *obstacles, *obstaclesCloud);
obstacles->clear();
obstaclesCloud->points[i] = cloud->at(obstacles->at(i));
if(flatObstaclesSet.size() == 0 ||
flatObstaclesSet.find(obstacles->at(i))==flatObstaclesSet.end())
{
obstaclesCloudWithoutFlatSurfaces->points[oi] = obstaclesCloud->points[i];
obstaclesCloudWithoutFlatSurfaces->points[oi].z = 0;
++oi;
}
}
// Part 2: segment floor and obstacles far from the robot
rtabmap::util3d::segmentObstaclesFromGround<pcl::PointXYZ>(
originalCloud_far,
ground,
obstacles,
normalKSearch_,
2.*groundNormalAngle_,
3.*clusterRadius_,
minClusterSize_,
segmentFlatObstacles_,
maxGroundHeight_);
if(groundPub_.getNumSubscribers() && ground.get() && ground->size())
{
pcl::PointCloud<pcl::PointXYZ>::Ptr groundCloud2 (new pcl::PointCloud<pcl::PointXYZ>);
pcl::copyPointCloud(*originalCloud_far, *ground, *groundCloud2);
*groundCloud += *groundCloud2;
}
if((obstaclesPub_.getNumSubscribers() || projObstaclesPub_.getNumSubscribers()) && obstacles.get() && obstacles->size())
{
pcl::PointCloud<pcl::PointXYZ>::Ptr obstacles2(new pcl::PointCloud<pcl::PointXYZ>);
pcl::copyPointCloud(*originalCloud_far, *obstacles, *obstacles2);
*obstaclesCloud += *obstacles2;
}
obstaclesCloudWithoutFlatSurfaces->resize(oi);
}
if(!localTransform.isIdentity())
@@ -346,16 +376,10 @@ private:
private:
std::string frameId_;
int normalKSearch_;
double groundNormalAngle_;
double clusterRadius_;
int minClusterSize_;
double maxObstaclesHeight_;
double maxGroundHeight_;
bool segmentFlatObstacles_;
std::string mapFrameId_;
bool waitForTransform_;
bool optimizeForCloseObjects_;
double projVoxelSize_;
rtabmap::OccupancyGrid grid_;
tf::TransformListener tfListener_;
src/nodelets/obstacles_detection_indoor.cpp
-267
View File
@@ -1,267 +0,0 @@
/*
Copyright (c) 2010-2016, Mathieu Labbe - IntRoLab - Universite de Sherbrooke
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 Universite de Sherbrooke 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 HOLDER 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.
*/
#include <ros/ros.h>
#include <pluginlib/class_list_macros.h>
#include <nodelet/nodelet.h>
#include <pcl/point_cloud.h>
#include <pcl/point_types.h>
#include <pcl/common/transforms.h>
#include <pcl_conversions/pcl_conversions.h>
#include <tf/transform_listener.h>
#include <sensor_msgs/PointCloud2.h>
#include <rtabmap_ros/MsgConversion.h>
#include "rtabmap/core/util3d_filtering.h"
#include "rtabmap/core/util3d_transforms.h"
namespace rtabmap_ros
{
class ObstaclesDetectionIndoor : public nodelet::Nodelet
{
public:
ObstaclesDetectionIndoor() :
frameId_(""),
minObstaclesHeight_(0.0), // if >0.0 -> disabled
maxObstaclesHeight_(0.0), // if <=0.0 -> disabled
minGroundHeight_(-0.05),
maxGroundHeight_(0.05),
waitForTransform_(false),
projVoxelSize_(0.01),
noiseFilterRadius_(0.0), // if<=0.0 -> disabled
noiseFilterMinNeighbors_(5)
{}
virtual ~ObstaclesDetectionIndoor()
{}
private:
virtual void onInit()
{
ros::NodeHandle & nh = getNodeHandle();
ros::NodeHandle & pnh = getPrivateNodeHandle();
int queueSize = 10;
pnh.param("queue_size", queueSize, queueSize);
pnh.param("frame_id", frameId_, frameId_);
pnh.param("min_obstacles_height", minObstaclesHeight_, minObstaclesHeight_);
pnh.param("max_obstacles_height", maxObstaclesHeight_, maxObstaclesHeight_);
pnh.param("min_ground_height", minGroundHeight_, minGroundHeight_);
pnh.param("max_ground_height", maxGroundHeight_, maxGroundHeight_);
pnh.param("noise_filter_radius", noiseFilterRadius_, noiseFilterRadius_);
pnh.param("noise_filter_min_neighbors", noiseFilterMinNeighbors_, noiseFilterMinNeighbors_);
pnh.param("wait_for_transform", waitForTransform_, waitForTransform_);
pnh.param("proj_voxel_size", projVoxelSize_, projVoxelSize_);
cloudSub_ = nh.subscribe("cloud", 1, &ObstaclesDetectionIndoor::callback, this);
groundPub_ = nh.advertise<sensor_msgs::PointCloud2>("ground", 1);
obstaclesPub_ = nh.advertise<sensor_msgs::PointCloud2>("obstacles", 1);
projObstaclesPub_ = nh.advertise<sensor_msgs::PointCloud2>("proj_obstacles", 1);
}
void callback(const sensor_msgs::PointCloud2ConstPtr & cloudMsg)
{
ros::WallTime time = ros::WallTime::now();
if (groundPub_.getNumSubscribers() == 0 && obstaclesPub_.getNumSubscribers() == 0 && projObstaclesPub_.getNumSubscribers() == 0)
{
// no one wants the results
return;
}
rtabmap::Transform localTransform;
try
{
if(waitForTransform_)
{
if(!tfListener_.waitForTransform(frameId_, cloudMsg->header.frame_id, cloudMsg->header.stamp, ros::Duration(1)))
{
NODELET_ERROR("Could not get transform from %s to %s after 1 second!", frameId_.c_str(), cloudMsg->header.frame_id.c_str());
return;
}
}
tf::StampedTransform tmp;
tfListener_.lookupTransform(frameId_, cloudMsg->header.frame_id, cloudMsg->header.stamp, tmp);
localTransform = rtabmap_ros::transformFromTF(tmp);
}
catch(tf::TransformException & ex)
{
NODELET_ERROR("%s",ex.what());
return;
}
pcl::PointCloud<pcl::PointXYZ>::Ptr originalCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::fromROSMsg(*cloudMsg, *originalCloud);
pcl::PointCloud<pcl::PointXYZ>::Ptr obstaclesCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::PointCloud<pcl::PointXYZ>::Ptr groundCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::PointCloud<pcl::PointXYZ>::Ptr projectedObstaclesCloud(new pcl::PointCloud<pcl::PointXYZ>);
if(originalCloud->size())
{
originalCloud = rtabmap::util3d::transformPointCloud(originalCloud, localTransform);
pcl::IndicesPtr indices(new std::vector<int>);
indices->resize(originalCloud->size());
for(unsigned int i=0; i<indices->size(); ++i)
{
indices->at(i) = i;
}
// segmentation
pcl::IndicesPtr groundIndices = rtabmap::util3d::passThrough(originalCloud, indices, "z", minGroundHeight_, maxGroundHeight_);
pcl::IndicesPtr obstacleIndices = rtabmap::util3d::extractIndices(originalCloud, groundIndices, true);
if(minObstaclesHeight_ <= 0.0 || maxObstaclesHeight_ > 0.0)
{
// std::numeric_limits<float>::lowest() exists only for c++11
obstacleIndices = rtabmap::util3d::passThrough(originalCloud, obstacleIndices, "z",
minObstaclesHeight_>0.0?std::numeric_limits<int>::min():minObstaclesHeight_,
maxObstaclesHeight_<=0.0?std::numeric_limits<int>::max():maxObstaclesHeight_);
}
// Do optional radius filtering to remove some noise
if(noiseFilterRadius_ > 0.0 && noiseFilterMinNeighbors_ > 0)
{
if(groundIndices->size())
{
groundIndices = rtabmap::util3d::radiusFiltering(originalCloud, groundIndices, noiseFilterRadius_, noiseFilterMinNeighbors_);
}
if(obstacleIndices->size())
{
obstacleIndices = rtabmap::util3d::radiusFiltering(originalCloud, obstacleIndices, noiseFilterRadius_, noiseFilterMinNeighbors_);
}
}
if(projObstaclesPub_.getNumSubscribers() && obstacleIndices->size())
{
projectedObstaclesCloud->resize(obstacleIndices->size());
for(unsigned int i=0; i<obstacleIndices->size(); ++i)
{
projectedObstaclesCloud->points[i] = originalCloud->at(obstacleIndices->at(i));
projectedObstaclesCloud->points[i].z = 0;
}
if(projVoxelSize_ > 0.0)
{
projectedObstaclesCloud = rtabmap::util3d::voxelize(projectedObstaclesCloud, projVoxelSize_);
}
}
if(!localTransform.isIdentity())
{
//transform back in topic frame
rtabmap::Transform localTransformInv = localTransform.inverse();
if(groundIndices->size())
{
pcl::transformPointCloud(*originalCloud, *groundIndices, *groundCloud, localTransformInv.toEigen3f());
}
if(obstacleIndices->size())
{
pcl::transformPointCloud(*originalCloud, *obstacleIndices, *obstaclesCloud, localTransformInv.toEigen3f());
}
}
else
{
if(groundIndices->size())
{
pcl::copyPointCloud(*originalCloud, *groundIndices, *groundCloud);
}
if(obstacleIndices->size())
{
pcl::copyPointCloud(*originalCloud, *obstacleIndices, *obstaclesCloud);
}
}
}
if(groundPub_.getNumSubscribers())
{
sensor_msgs::PointCloud2 rosCloud;
pcl::toROSMsg(*groundCloud, rosCloud);
rosCloud.header = cloudMsg->header;
//publish the message
groundPub_.publish(rosCloud);
}
if(obstaclesPub_.getNumSubscribers())
{
sensor_msgs::PointCloud2 rosCloud;
pcl::toROSMsg(*obstaclesCloud, rosCloud);
rosCloud.header = cloudMsg->header;
//publish the message
obstaclesPub_.publish(rosCloud);
}
if(projObstaclesPub_.getNumSubscribers())
{
sensor_msgs::PointCloud2 rosCloud;
pcl::toROSMsg(*projectedObstaclesCloud, rosCloud);
rosCloud.header.stamp = cloudMsg->header.stamp;
rosCloud.header.frame_id = frameId_;
//publish the message
projObstaclesPub_.publish(rosCloud);
}
NODELET_DEBUG("Obstacle segmentation time = %f s", (ros::WallTime::now() - time).toSec());
}
private:
std::string frameId_;
double minObstaclesHeight_;
double maxObstaclesHeight_;
double minGroundHeight_;
double maxGroundHeight_;
bool waitForTransform_;
double projVoxelSize_;
double noiseFilterRadius_;
int noiseFilterMinNeighbors_;
tf::TransformListener tfListener_;
ros::Publisher groundPub_;
ros::Publisher obstaclesPub_;
ros::Publisher projObstaclesPub_;
ros::Subscriber cloudSub_;
};
PLUGINLIB_EXPORT_CLASS(rtabmap_ros::ObstaclesDetectionIndoor, nodelet::Nodelet);
}
src/nodelets/obstacles_detection_old.cpp
+357
View File
@@ -0,0 +1,357 @@
/*
Copyright (c) 2010-2016, Mathieu Labbe - IntRoLab - Universite de Sherbrooke
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 Universite de Sherbrooke 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 HOLDER 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.
*/
#include <ros/ros.h>
#include <pluginlib/class_list_macros.h>
#include <nodelet/nodelet.h>
#include <pcl/point_cloud.h>
#include <pcl/point_types.h>
#include <pcl_conversions/pcl_conversions.h>
#include <tf/transform_listener.h>
#include <sensor_msgs/PointCloud2.h>
#include <rtabmap_ros/MsgConversion.h>
#include "rtabmap/core/util3d.h"
#include "rtabmap/core/util3d_filtering.h"
#include "rtabmap/core/util3d_mapping.h"
#include "rtabmap/core/util3d_transforms.h"
namespace rtabmap_ros
{
class ObstaclesDetectionOld : public nodelet::Nodelet
{
public:
ObstaclesDetectionOld() :
frameId_("base_link"),
normalKSearch_(20),
groundNormalAngle_(M_PI_4),
clusterRadius_(0.05),
minClusterSize_(20),
maxObstaclesHeight_(0.0), // if<=0.0 -> disabled
maxGroundHeight_(0.0), // if<=0.0 -> disabled, used only if detect_flat_obstacles is true
segmentFlatObstacles_(false),
waitForTransform_(false),
optimizeForCloseObjects_(false),
projVoxelSize_(0.01)
{}
virtual ~ObstaclesDetectionOld()
{}
private:
virtual void onInit()
{
ros::NodeHandle & nh = getNodeHandle();
ros::NodeHandle & pnh = getPrivateNodeHandle();
int queueSize = 10;
pnh.param("queue_size", queueSize, queueSize);
pnh.param("frame_id", frameId_, frameId_);
pnh.param("normal_k", normalKSearch_, normalKSearch_);
pnh.param("ground_normal_angle", groundNormalAngle_, groundNormalAngle_);
if(pnh.hasParam("normal_estimation_radius") && !pnh.hasParam("cluster_radius"))
{
NODELET_WARN("Parameter \"normal_estimation_radius\" has been renamed "
"to \"cluster_radius\"! Your value is still copied to "
"corresponding parameter. Instead of normal radius, nearest neighbors count "
"\"normal_k\" is used instead (default 20).");
pnh.param("normal_estimation_radius", clusterRadius_, clusterRadius_);
}
else
{
pnh.param("cluster_radius", clusterRadius_, clusterRadius_);
}
pnh.param("min_cluster_size", minClusterSize_, minClusterSize_);
pnh.param("max_obstacles_height", maxObstaclesHeight_, maxObstaclesHeight_);
pnh.param("max_ground_height", maxGroundHeight_, maxGroundHeight_);
pnh.param("detect_flat_obstacles", segmentFlatObstacles_, segmentFlatObstacles_);
pnh.param("wait_for_transform", waitForTransform_, waitForTransform_);
pnh.param("optimize_for_close_objects", optimizeForCloseObjects_, optimizeForCloseObjects_);
pnh.param("proj_voxel_size", projVoxelSize_, projVoxelSize_);
cloudSub_ = nh.subscribe("cloud", 1, &ObstaclesDetectionOld::callback, this);
groundPub_ = nh.advertise<sensor_msgs::PointCloud2>("ground", 1);
obstaclesPub_ = nh.advertise<sensor_msgs::PointCloud2>("obstacles", 1);
projObstaclesPub_ = nh.advertise<sensor_msgs::PointCloud2>("proj_obstacles", 1);
}
void callback(const sensor_msgs::PointCloud2ConstPtr & cloudMsg)
{
ros::WallTime time = ros::WallTime::now();
if (groundPub_.getNumSubscribers() == 0 && obstaclesPub_.getNumSubscribers() == 0 && projObstaclesPub_.getNumSubscribers() == 0)
{
// no one wants the results
return;
}
rtabmap::Transform localTransform;
try
{
if(waitForTransform_)
{
if(!tfListener_.waitForTransform(frameId_, cloudMsg->header.frame_id, cloudMsg->header.stamp, ros::Duration(1)))
{
NODELET_ERROR("Could not get transform from %s to %s after 1 second!", frameId_.c_str(), cloudMsg->header.frame_id.c_str());
return;
}
}
tf::StampedTransform tmp;
tfListener_.lookupTransform(frameId_, cloudMsg->header.frame_id, cloudMsg->header.stamp, tmp);
localTransform = rtabmap_ros::transformFromTF(tmp);
}
catch(tf::TransformException & ex)
{
NODELET_ERROR("%s",ex.what());
return;
}
pcl::PointCloud<pcl::PointXYZ>::Ptr originalCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::fromROSMsg(*cloudMsg, *originalCloud);
//Common variables for all strategies
pcl::IndicesPtr ground, obstacles;
pcl::PointCloud<pcl::PointXYZ>::Ptr obstaclesCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::PointCloud<pcl::PointXYZ>::Ptr groundCloud(new pcl::PointCloud<pcl::PointXYZ>);
pcl::PointCloud<pcl::PointXYZ>::Ptr obstaclesCloudWithoutFlatSurfaces(new pcl::PointCloud<pcl::PointXYZ>);
if(originalCloud->size())
{
originalCloud = rtabmap::util3d::transformPointCloud(originalCloud, localTransform);
if(maxObstaclesHeight_ > 0)
{
// std::numeric_limits<float>::lowest() exists only for c++11
originalCloud = rtabmap::util3d::passThrough(originalCloud, "z", std::numeric_limits<int>::min(), maxObstaclesHeight_);
}
if(originalCloud->size())
{
if(!optimizeForCloseObjects_)
{
// This is the default strategy
pcl::IndicesPtr flatObstacles(new std::vector<int>);
rtabmap::util3d::segmentObstaclesFromGround<pcl::PointXYZ>(
originalCloud,
ground,
obstacles,
normalKSearch_,
groundNormalAngle_,
clusterRadius_,
minClusterSize_,
segmentFlatObstacles_,
maxGroundHeight_,
&flatObstacles);
if(groundPub_.getNumSubscribers() &&
ground.get() && ground->size())
{
pcl::copyPointCloud(*originalCloud, *ground, *groundCloud);
}
if((obstaclesPub_.getNumSubscribers() || projObstaclesPub_.getNumSubscribers()) &&
obstacles.get() && obstacles->size())
{
// remove flat obstacles from obstacles
std::set<int> flatObstaclesSet;
if(projObstaclesPub_.getNumSubscribers())
{
flatObstaclesSet.insert(flatObstacles->begin(), flatObstacles->end());
}
obstaclesCloud->resize(obstacles->size());
obstaclesCloudWithoutFlatSurfaces->resize(obstacles->size());
int oi=0;
for(unsigned int i=0; i<obstacles->size(); ++i)
{
obstaclesCloud->points[i] = originalCloud->at(obstacles->at(i));
if(flatObstaclesSet.size() == 0 ||
flatObstaclesSet.find(obstacles->at(i))==flatObstaclesSet.end())
{
obstaclesCloudWithoutFlatSurfaces->points[oi] = obstaclesCloud->points[i];
obstaclesCloudWithoutFlatSurfaces->points[oi].z = 0;
++oi;
}
}
obstaclesCloudWithoutFlatSurfaces->resize(oi);
if(obstaclesCloudWithoutFlatSurfaces->size() && projVoxelSize_ > 0.0)
{
obstaclesCloudWithoutFlatSurfaces = rtabmap::util3d::voxelize(obstaclesCloudWithoutFlatSurfaces, projVoxelSize_);
}
}
}
else
{
// in this case optimizeForCloseObject_ is true:
// we divide the floor point cloud into two subsections, one for all potential floor points up to 1m
// one for potential floor points further away than 1m.
// For the points at closer range, we use a smaller normal estimation radius and ground normal angle,
// which allows to detect smaller objects, without increasing the number of false positive.
// For all other points, we use a bigger normal estimation radius (* 3.) and tolerance for the
// grond normal angle (* 2.).
pcl::PointCloud<pcl::PointXYZ>::Ptr originalCloud_near = rtabmap::util3d::passThrough(originalCloud, "x", std::numeric_limits<int>::min(), 1.);
pcl::PointCloud<pcl::PointXYZ>::Ptr originalCloud_far = rtabmap::util3d::passThrough(originalCloud, "x", 1., std::numeric_limits<int>::max());
// Part 1: segment floor and obstacles near the robot
rtabmap::util3d::segmentObstaclesFromGround<pcl::PointXYZ>(
originalCloud_near,
ground,
obstacles,
normalKSearch_,
groundNormalAngle_,
clusterRadius_,
minClusterSize_,
segmentFlatObstacles_,
maxGroundHeight_);
if(groundPub_.getNumSubscribers() && ground.get() && ground->size())
{
pcl::copyPointCloud(*originalCloud_near, *ground, *groundCloud);
ground->clear();
}
if((obstaclesPub_.getNumSubscribers() || projObstaclesPub_.getNumSubscribers()) && obstacles.get() && obstacles->size())
{
pcl::copyPointCloud(*originalCloud_near, *obstacles, *obstaclesCloud);
obstacles->clear();
}
// Part 2: segment floor and obstacles far from the robot
rtabmap::util3d::segmentObstaclesFromGround<pcl::PointXYZ>(
originalCloud_far,
ground,
obstacles,
normalKSearch_,
2.*groundNormalAngle_,
3.*clusterRadius_,
minClusterSize_,
segmentFlatObstacles_,
maxGroundHeight_);
if(groundPub_.getNumSubscribers() && ground.get() && ground->size())
{
pcl::PointCloud<pcl::PointXYZ>::Ptr groundCloud2 (new pcl::PointCloud<pcl::PointXYZ>);
pcl::copyPointCloud(*originalCloud_far, *ground, *groundCloud2);
*groundCloud += *groundCloud2;
}
if((obstaclesPub_.getNumSubscribers() || projObstaclesPub_.getNumSubscribers()) && obstacles.get() && obstacles->size())
{
pcl::PointCloud<pcl::PointXYZ>::Ptr obstacles2(new pcl::PointCloud<pcl::PointXYZ>);
pcl::copyPointCloud(*originalCloud_far, *obstacles, *obstacles2);
*obstaclesCloud += *obstacles2;
}
}
if(!localTransform.isIdentity())
{
//transform back in topic frame
rtabmap::Transform localTransformInv = localTransform.inverse();
if(groundCloud->size())
{
groundCloud = rtabmap::util3d::transformPointCloud(groundCloud, localTransformInv);
}
if(obstaclesCloud->size())
{
obstaclesCloud = rtabmap::util3d::transformPointCloud(obstaclesCloud, localTransformInv);
}
}
}
}
if(groundPub_.getNumSubscribers())
{
sensor_msgs::PointCloud2 rosCloud;
pcl::toROSMsg(*groundCloud, rosCloud);
rosCloud.header = cloudMsg->header;
//publish the message
groundPub_.publish(rosCloud);
}
if(obstaclesPub_.getNumSubscribers())
{
sensor_msgs::PointCloud2 rosCloud;
pcl::toROSMsg(*obstaclesCloud, rosCloud);
rosCloud.header = cloudMsg->header;
//publish the message
obstaclesPub_.publish(rosCloud);
}
if(projObstaclesPub_.getNumSubscribers())
{
sensor_msgs::PointCloud2 rosCloud;
pcl::toROSMsg(*obstaclesCloudWithoutFlatSurfaces, rosCloud);
rosCloud.header.stamp = cloudMsg->header.stamp;
rosCloud.header.frame_id = frameId_;
//publish the message
projObstaclesPub_.publish(rosCloud);
}
NODELET_DEBUG("Obstacles segmentation time = %f s", (ros::WallTime::now() - time).toSec());
}
private:
std::string frameId_;
int normalKSearch_;
double groundNormalAngle_;
double clusterRadius_;
int minClusterSize_;
double maxObstaclesHeight_;
double maxGroundHeight_;
bool segmentFlatObstacles_;
bool waitForTransform_;
bool optimizeForCloseObjects_;
double projVoxelSize_;
tf::TransformListener tfListener_;
ros::Publisher groundPub_;
ros::Publisher obstaclesPub_;
ros::Publisher projObstaclesPub_;
ros::Subscriber cloudSub_;
};
PLUGINLIB_EXPORT_CLASS(rtabmap_ros::ObstaclesDetectionOld, nodelet::Nodelet);
}