Hello, I've been working with a turtlebot with a kobuki base , i'm trying to use robot_pose_ekf but i found a tf problem,
in this case base_footprint will have 2 parents (odom_combined published by robot_pose_ekf node and odom published by
nodelet_manager node)
So how can i resolve this problem, should i change some parameters in launch files?
Any help please
↧
problem with tf tree, robot_pose_ekf, odom_combined
↧
Calculating with TF/TF2
Hi,
Currently I have a two interactive markers in the map frame.
Now I want to the coordinates of a point on the line between them with a distance x from one of those points.
Can TF/TF2 do that, if yes, how?
Or do I have to do that myself?
Do I have to publish tf messages or can I calculate with non-published transforms internally somehow?
↧
↧
Does tf.lookupTransform (python) take time to fire up?
Hi there,
I have an issue I don't understand with tf in Python. Consider this code:
import rospy, tf
rospy.init_node('test')
l = tf.TransformListener()
# rospy.sleep(0.5)
# l.waitForTransform('BODY', 'manipulation_marker', rospy.Time(0), rospy.Time(10))
print l.lookupTransform('BODY', 'manipulation_marker', rospy.Time(0))
When I run it, I get a LookupException:
LookupException: "BODY" passed to lookupTransform argument target_frame does not exist.
But if I uncomment the sleep() before the lookupTransform, it works fine.
So, I was thinking I could use the waitForTransform() function to replace this sleep() and get the transform as soon as it is available. (Next commented line.) But it doesn't work either and I get a generic Exception() with no error message!
To give you more information about my TF, here is the output from tf_monitor:
RESULTS: for BODY to manipulation_marker
Chain is: * -> * -> * -> BODY
Net delay avg = 0.0424558: max = 0.0998758
All Broadcasters:
Node: unknown_publisher 111.186 Hz, Average Delay: -0.221831 Max Delay: 0.0471964
What should I do here to get the transform as soon as it is available?
↧
TF tutorial crash
I'm following this tutorial
http://www.ros.org/wiki/tf/Tutorials/Introduction%20to%20tf
The last step in the launch file
↧
problem with hector slam
HI All,
I'm hoping someone might be able to help. I've been trying to get gmapping working ut the update rates was incredibly slow (~5 seconds from moving the robot to the map moving) so I'm giving the hector slam ago.
This problem must be due to the tf I'm publishing, I'm just not sure what hector slam actually needs me to publish vs what is will publish for itself.
I'm starting it with the standard (i.e. I've changed nothing):
roslaunch hector_slam_launch tutorial.launch
I'm have made the tf as simple as possible
And my system graph looks like:
Hector slam doesn't seem to be doing anything. All I get printed out is:
process[rviz-1]: started with pid [8374]
process[hector_mapping-2]: started with pid [8375]
process[hector_trajectory_server-3]: started with pid [8379]
process[hector_geotiff_node-4]: started with pid [8383]
[ INFO] [1429946593.528284272]: Waiting for tf transform data between frames /map and scanmatcher_frame to become available
[ INFO] [1429946594.644483794]: Successfully initialized hector_geotiff MapWriter plugin TrajectoryMapWriter.
[ INFO] [1429946594.644839942]: Geotiff node started
HectorSM map lvl 0: cellLength: 0.05 res x:2048 res y: 2048
HectorSM map lvl 1: cellLength: 0.1 res x:1024 res y: 1024
[ INFO] [1429946598.082143817]: HectorSM p_base_frame_: base_footprint
[ INFO] [1429946598.082288732]: HectorSM p_map_frame_: map
[ INFO] [1429946598.082325095]: HectorSM p_odom_frame_: nav
[ INFO] [1429946598.082375357]: HectorSM p_scan_topic_: scan
[ INFO] [1429946598.082410037]: HectorSM p_use_tf_scan_transformation_: true
[ INFO] [1429946598.082460158]: HectorSM p_pub_map_odom_transform_: true
[ INFO] [1429946598.082495348]: HectorSM p_scan_subscriber_queue_size_: 5
[ INFO] [1429946598.082536563]: HectorSM p_map_pub_period_: 2.000000
[ INFO] [1429946598.082575411]: HectorSM p_update_factor_free_: 0.400000
[ INFO] [1429946598.082610219]: HectorSM p_update_factor_occupied_: 0.900000
[ INFO] [1429946598.082670763]: HectorSM p_map_update_distance_threshold_: 0.400000
[ INFO] [1429946598.082713224]: HectorSM p_map_update_angle_threshold_: 0.060000
[ INFO] [1429946598.082752979]: HectorSM p_laser_z_min_value_: -1.000000
[ INFO] [1429946598.082792572]: HectorSM p_laser_z_max_value_: 1.000000
[ WARN] [1429946613.537432314]: No transform between frames /map and scanmatcher_frame available after 20.009908 seconds of waiting. This warning only prints once.
If anyone could advise on what I need to correct I'd be most grateful
Kind Regards
Mark
↧
↧
tranformation using static publisher and listener
Hi,
If I have two frames I want to do transformation.. in my code i have the following:
void laserCallback (const sensor_msgs::LaserScan::ConstPtr& scan_in)
{
double t1 = ros::Time::now().toSec() ;
//std::cout<<"LASER" << scan_in->header.frame_id << std::endl ;
if(!listener_.waitForTransform(scan_in->header.frame_id,
"/uav/baselink_ENU",
scan_in->header.stamp + ros::Duration().fromSec(scan_in->ranges.size()*scan_in->time_increment),
ros::Duration(1.0)))
{
std::cout << "RETURN" << std::endl ;
return;
}
sensor_msgs::PointCloud msg;
// Which one of the follwoing should I use ?
//projector_.projectLaser(*scan_in, msg);
//projector_.transformLaserScanToPointCloud("/uav/baselink_ENU",*scan_in, msg,listener_);
}
Is this part needed if I have a static transformation :
if(!listener_.waitForTransform(scan_in->header.frame_id,
"/uav/baselink_ENU",
scan_in->header.stamp + ros::Duration().fromSec(scan_in->ranges.size()*scan_in->time_increment),
ros::Duration(1.0)))
{
std::cout << "RETURN" << std::endl ;
return;
}
I mean is it enough to have a static tf publisher and the following line ?
projector_.transformLaserScanToPointCloud("/uav/baselink_ENU",*scan_in, msg,listener_);
↧
transform laser data to pointcloud usig tf transform without static transformation
Hello,
I have a laser that I connected it up-side-down with the quadrotor. The quadrotor system gives me the child base_link frame to the parent world frame.
When I run hokyou node, the /scan topic which has a laserScan data that are registered with laser frame as a frame id.
Now when I read the laser data from the topic /scan and I use the following transformation:
void laserCallback (const sensor_msgs::LaserScan::ConstPtr& scan_in)
{
double t1 = ros::Time::now().toSec() ;
//std::cout<<"LASER" << scan_in->header.frame_id << std::endl ;
if(!listener_.waitForTransform(scan_in->header.frame_id,
"/uav/baselink_ENU",
scan_in->header.stamp + ros::Duration().fromSec(scan_in->ranges.size()*scan_in->time_increment),
ros::Duration(1.0)))
{
std::cout << "RETURN" << std::endl ;
return;
}
sensor_msgs::PointCloud msg;
//projector_.projectLaser(*scan_in, msg);
projector_.transformLaserScanToPointCloud("/uav/baselink_ENU",*scan_in, msg,listener_);
}
Does this directly transform the scaned data into xs and ys to the base_link frame or I should use static transformation?
↧
urdf in rviz is too deep
Hi, I setup my robot system and its working good.
I am using hector mapping with urdf and a laserscanner and I am getting the following result:
I dont know why exactly the robot is kind of swimming in the world, here is my tf tree:
Another Question is now, how can I make the Robot move autonomously and how can I integrate a path planner?
What is a common used module to make the robot move and avoid obstacles and crashes?
thanks in advance, felix
↧
Gmapping and TF frames
Hello,
I'm using Ubuntu 14.04 and ROS indigo
I published base_link>camera_link frames using static_transform_publisher as follows:
↧
↧
Using the estimated robot pose from robot_pose_ekf with amcl
Hi all,
I am using [robot_pose_ekf](http://wiki.ros.org/robot_pose_ekf) to fuse data from Wheel encoders and IMU to estimate the 3D pose of the robot. Then, I am using [amcl](http://wiki.ros.org/amcl) to take care of odometry drift and finally the [move_base](http://wiki.ros.org/move_base) package does the navigation. I am generating the odom message of the type `nav_msgs/Odometry` from the wheel encoders which is used by the [base_local_planner](http://wiki.ros.org/base_local_planner?distro=indigo). I know robot_pose_ekf provides odom_combined to base_link transformation which is used by amcl.
My question is once robot_pose_ekf fuses the data from wheel encoders and IMU and generates the estimated 3D pose of the robot of the type `geometry_msgs/PoseWithCovarianceStamped` on the topic `robot_pose_ekf/odom_combined`, how and where is this output used in the navigation package as a whole (move_base or amcl)?
Thanks in advance.
Naman Kumar
↧
robot_base_frame parameter in move_base
Hi all,
I have a mobile robot with IMU, wheel encoders and hokuyo laser. I am following [this](http://answers.ros.org/question/11682/robot_pose_ekf-with-an-external-sensor/) to set up the transformations and all the frames. I am using robot_pose_ekf to fuse data from IMU and wheel encoders and get the `odom_combined -> base_footprint` transformation. As suggested in the above link, in wheel_odometry, `odom_combined` is the header frame id and `base_footprint` is the child_frame_id and I have a fixed transformation from `base_footprint->base_link` . Now my question is what should be the robot_base_frame parameter in `global_costmap_params.yaml` and `local_costmap_params.yaml` in move_base? Should it be base_link(this is the default) or base_footprint and why?
Thanks in advance.
Naman Kumar
↧
base_link to base_footprint transform ?
Hi,
I am trying to use robot_pose_ekf for my navigation stack,
I have done every thing according to [this thread](http://answers.ros.org/question/2642/robot_pose_ekf-with-an-external-sensor)
but I faced a problem about my transformations!
Robot_pose_ekf is publishing a base_footprint tf, but i dont know how should i connect base_link tf to base_footprint tf. what is the relation between them?
↧
hector_slam where to put nav frame?
In mapping_default.launch, default "odom_frame" is "nav". And the hector_mapping node publishes transform from map to nav.
If I have put nav as a parent of base_footprint with 0,0,0,0,0 transformation, the mapping runs fine. But I don't understand the reasoning behind this.
Can somebody explain to me for calling odom_frame nav.
↧
↧
An error for publishing nav_msgs/Odometry
In a simulator, I am trying to write a python script which subscribe a nav_msgs/Odometry topic to add a frame between odom and base_footprint. It then also publishes another nav_msgs/Odometry topic under frame odom.
The callback function for the subscriber is here:
#!/usr/bin/env python
import roslib
roslib.load_manifest('simulator')
import rospy, tf
from geometry_msgs.msg import Point, Quaternion, PoseStamped
from nav_msgs.msg import Odometry
import xml.dom.minidom
def newDataAvailable(robotPose):
robotPosition=float(robotPose.pose.pose.position.x) -x, float(robotPose.pose.pose.position.y) -y,0.0
robotOrientation=robotPose.pose.pose.orientation.x, robotPose.pose.pose.orientation.y, robotPose.pose.pose.orientation.z, robotPose.pose.pose.orientation.w
robotVelocity=(robotPose.twist.twist.linear.x,robotPose.twist.twist.linear.y,robotPose.twist.twist.linear.z)
currentTime = rospy.Time.now()
br.sendTransform(robotPosition, robotOrientation, currentTime, "base_footprint", "odom")
odom = Odometry()
odom.header.stamp = rospy.Time.now()
odom.header.frame_id = parentFrameId
odom.child_frame_id = frameId
odom.pose.pose.position = robotPosition
print odom.pose.pose.position
odom.pose.pose.orientation = robotOrientation
print "print orientation"
print odom.pose.pose.orientation
odom.twist.twist.linear=Point(*robotVelocity)
print odom.twist
odomPublisher.publish(odom)
def main():
rospy.init_node('simulator_robot_tf_broadcaster')
global frameId, parentFrameId, br, x, y, odomPublisher
frameId = rospy.get_param("frame", "base_footprint")
parentFrameId = rospy.get_param("parent_frame", "odom")
robotgroundtruthTopic = "/robot_pose"
rospy.Subscriber(robotgroundtruthTopic, Odometry, newDataAvailable)
odomPublisher=rospy.Publisher("/base_pose_ground_truth", Odometry, 10)
scene_file_param=rospy.get_param("/scene_file", "scene.xml")
dom = xml.dom.minidom.parse(scene_file_param)
root = dom.documentElement
itemlist = dom.getElementsByTagName('agent')
for i in itemlist:
pn = i.getAttribute("type")
if pn=="2": # robot:2
print "I got the robot's initial postion"
x = i.getAttribute("x")
#print x
y = i.getAttribute("y")
#print y
br = tf.TransformBroadcaster()
x=float(x)
y=float(y)
print x,y
rospy.spin()
if __name__ == '__main__':
main()
When I run it, an error comes out:
File "/opt/ros/hydro/lib/python2.7/dist-packages/nav_msgs/msg/_Odometry.py", line 161, in serialize
buff.write(_struct_7d.pack(_x.pose.pose.position.x, _x.pose.pose.position.y, _x.pose.pose.position.z, _x.pose.pose.orientation.x, _x.pose.pose.orientation.y, _x.pose.pose.orientation.z, _x.pose.pose.orientation.w))
AttributeError: 'tuple' object has no attribute 'x'
All print functions work fine. However, no message outputs from topic "/base_pose_ground_truth"
EDIT1: I modified some places of code as:
robotPosition=robotPose.pose.pose.position.x - x, robotPose.pose.pose.position.y-y, 0.0
robotOrientation=(robotPose.pose.pose.orientation.x, robotPose.pose.pose.orientation.y, robotPose.pose.pose.orientation.z, robotPose.pose.pose.orientation.w)
robotVelocity=(robotPose.twist.twist.linear.x,robotPose.twist.twist.linear.y,robotPose.twist.twist.linear.z)
currentTime = rospy.Time.now()
br.sendTransform(robotPosition, robotOrientation, currentTime, "base_footprint", "odom")
odom = Odometry()
odom.header.stamp = rospy.Time.now()
odom.header.frame_id = parentFrameId
odom.child_frame_id = frameId
odom.pose.pose.position = Point(*robotPosition)
odom.pose.pose.orientation =Quaternion(*robotOrientation)
odom.twist.twist.linear=Point(*robotVelocity)
odomPublisher.publish(odom)
Then above error(simplified) disappears. Message output from topic "/base_pose_ground_truth" also but in a lower speed comparing to topic "/robot_pose"( why?)
However, as it is mentioned before, the above node adds a frame between "odom" and "base_footprint". Then this node is connected to [fakelocalization](http://wiki.ros.org/fake_localization).
But, the link between map and odom doesn't constructed by fakelocalization successfully.
And an warning comes out:
[WARN] [WallTime: 1430749249.205509] Inbound TCP/IP connection failed: 'int' object has no attribute 'peer_subscribe'
Please tell me the reason why. Thank you!
↧
Add a tf frame in a simulator
Hi, all. I want to add a frame between "odom" and "base_footprint" in a simulator by imitating [navigation_stage](http://wiki.ros.org/navigation_stage). Then it is connected to [fakelocalization](http://wiki.ros.org/fake_localization). In this way, the tf tree for this simulator is constructed.
It is a python script which subscribes a "/robot_pose" topic to compute the transform between odom and base_footprint, then it publishes another "/base_pose_ground_truth" topic under frame "/odom".
The codes of the script are listed below.
#!/usr/bin/env python
import roslib
roslib.load_manifest('simulator')
import rospy, tf
from geometry_msgs.msg import Point, Quaternion, PoseStamped
from nav_msgs.msg import Odometry
import xml.dom.minidom
def newDataAvailable(robotPose):
robotPosition=robotPose.pose.pose.position.x - x, robotPose.pose.pose.position.y-y, 0.0
robotOrientation=(robotPose.pose.pose.orientation.x, robotPose.pose.pose.orientation.y, robotPose.pose.pose.orientation.z, robotPose.pose.pose.orientation.w)
robotVelocity=(robotPose.twist.twist.linear.x,robotPose.twist.twist.linear.y,robotPose.twist.twist.linear.z)
currentTime = rospy.Time.now()
br.sendTransform(robotPosition, robotOrientation, currentTime, "base_footprint", "odom")
odom = Odometry()
odom.header.stamp = rospy.Time.now()
odom.header.frame_id = parentFrameId
odom.child_frame_id = frameId
odom.pose.pose.position = Point(*robotPosition)
odom.pose.pose.orientation =Quaternion(*robotOrientation)
odom.twist.twist.linear=Point(*robotVelocity)
odomPublisher.publish(odom)
def main():
rospy.init_node('simulator_robot_tf_broadcaster')
global frameId, parentFrameId, br, x, y, odomPublisher
frameId = rospy.get_param("frame", "base_footprint")
parentFrameId = rospy.get_param("parent_frame", "odom")
robotgroundtruthTopic = "/robot_pose"
rospy.Subscriber(robotgroundtruthTopic, Odometry, newDataAvailable)
odomPublisher=rospy.Publisher("/base_pose_ground_truth", Odometry, 1)
scene_file_param=rospy.get_param("/scene_file", "scene.xml")
dom = xml.dom.minidom.parse(scene_file_param)
root = dom.documentElement
itemlist = dom.getElementsByTagName('agent')
for i in itemlist:
pn = i.getAttribute("type")
if pn=="2": # robot:2
print "I got the robot's initial postion"
x = i.getAttribute("x")
#print x
y = i.getAttribute("y")
#print y
br = tf.TransformBroadcaster()
x=float(x)
y=float(y)
rospy.set_param("/fake_localization/delta_x",-x)
rospy.set_param("/fake_localization/delta_y",-y)
print x,y
rospy.spin()
if __name__ == '__main__':
main()
When I run it, all coordinates works as expected. However, there exists serious delay for computing tf. In rviz, the TF tree doesn't displayed in real-time. For example, the displayed coordinate "base_footprint" in rviz is "following" the robot ( I am quite sure that this is time delay, not the transform error in distance, because the displayed "base_footprint" axes can get the robot's position).
And an warning comes out all the time:
[WARN] [WallTime: 1430749249.205509] Inbound TCP/IP connection failed: 'int' object has no attribute 'peer_subscribe'
Another phenomenon is the message outputing from topic "/base_pose_ground_truth" is in a lower speed comparing to topic "/robot_pose"( why?)
Please tell me the reason why. Thank you!
↧
rviz don't follow the tf configuration of the .rviz file
Hello! My problem is that I want to display some of the TF and hide others. I can do it by selecting what I want by hand, but if I want to save it in a .rviz file and launching it with `rosrun rviz rviz -d my_config.rviz`, it display all my tf.
The .rviz file is well saved, here is an abstract of the /TF class in the my_config.rviz :
Visualization Manager:
Class: ""
Displays:
- Alpha: 0.5
Cell Size: 1
Class: rviz/Grid
Color: 160; 160; 164
Enabled: true
Line Style:
Line Width: 0.03
Value: Lines
Name: Grid
Normal Cell Count: 0
Offset:
X: 0
Y: 0
Z: 0
Plane: XY
Plane Cell Count: 10
Reference Frame:
Value: true
- Class: rviz/TF
Enabled: true
Frame Timeout: 15
Frames:
All Enabled: false
WheelFL_link:
Value: false
WheelFR_link:
Value: false
ar_marker_1:
Value: true
"my_config.rviz" is the right one loaded ( I tried to add and remove a display type to be sure ), and I am using the rviz version 1.11.7 (indigo), Compiled against OGRE version 1.8.1 (Byatis).
↧
Problem mixing lidar bag with live tf
During development, I am trying to replay lidar data from a bag file and combine it with simulated new robot positions, generated during the simulation run. The lidar data I recorded a few weeks ago, and the simulated positions are computed on the fly by a controller and command based odometry nodes.
However there is some problem with tf occuring, and causing assemble_scans to give no output. Assemble_scans works fine when I replay tf's recorded inside the bag together with the lidar; but not when I filter them out of the bag file and play my live versions. (I'm running in simtime mode.)
I found a couple of questions and answers regarding tf but nothing definitive yet. From these: is this a bug/limitation of tf itself, where it's not using ros time and using wallclock time instead? If so, is anyone working on fixing it, and are there any current workarounds? Or if not, any ideas what else could be going wrong here?
↧
↧
RGBDSLAM - Slow transform publishing
Hi there,
I have the RGBD SLAM v2 package (http://felixendres.github.io/rgbdslam_v2/) running on a Core i7 computer with a Nvidia GTX 780 graphics card. The computer runs Ubuntu 12.04.5 and ROS Hydro.
I want to use the package for robot localization, but it seems like I'm getting a too low pub rate of the transform.
In my launch file I have:
↧
For frame[base_link]: No transform to fixed frame using IMU
Hi all,
I have a mobile robot and I would like it to navigate around a room and I already have a map of the room. I am using rotary encoders for odometry. I am using [robot_pose_ekf](http://wiki.ros.org/robot_pose_ekf) to merge data from rotary encoders and IMU. I am using [amcl](http://wiki.ros.org/amcl) for localization and [move_base](http://wiki.ros.org/move_base) for planning. Now the problem is, when I set the fixed frame in RVIZ as map, the status of IMU keeps on oscillating from `Status:OK` to following:
The tf tree is shown which looks good to me:
[C:\fakepath\tf_tree.png](/upfiles/14313589068939156.png)
**IMU part of the Code**
For IMU, I am using http://wiki.ros.org/phidgets_imu and http://wiki.ros.org/imu_filter_madgwick.
↧
I have a question about URDF models and tf transformations
Hi guys,
I was convinced that if one defines, let's say 2 links, in a urdf file and joins them toghether with a joint element, then the physic relations between both links is going to be recognized automatically from tf when reading (parsing) the file at the moment to start the program. So it would be possible - I thought - to listen that transformation in my C++ program.
To make it clearer:
I have the following urdf file (of course it is just a barebone example)
##### base_link ##############################################################
↧