James Diprose's profile - activity

Hi

I'm a bit confused since I'm new to ROS. I'm using the openni_tracker and ar_kinect packages together. Rviz complains that it can't transform from any of the skeleton topics (e.g. /head1) to /camera_depth_optical_frame (which is the fixed frame).

So I'm wondering how do I transform from /openni_depth_frame to /camera_link? /openni_depth_frame is the root of the tf tree for the skeleton produced by openni_tracker and /camera_link is the root of the tf tree for ar_kinect.

This is the output from "rosrun tf view_frames" so you can see the tf's that are running.

Any help would be much appreciated :)

James

No worries :)

Its because the launch file in openni_launch hasn't been updated to remove the slashes from the cameras frame_ids. You can edit the launch file openni.launch and remove the "/" from "/camera_rgb_optical_frame" (I cant remember exactly, but I think that launch file actually includes another file with a whole lot of parameters - which is the one you want to edit).

So I'm going to try just with K which has the variables for the raw images that the softkinetic driver provides...

Hey Corot! Thanks for your answer. Yeah I was using the equations that you would use to rectify an image (I think) by using the variables from P in camera info, but it always seemed like the transformed (u,v) points didn't match up correctly with the point cloud.

Hi

I'm using a Creative Senz3D for human-robot interaction perception tasks. I want to project an image point (u,v) into world coordinates, but where there is no corresponding point in the point cloud. E.g. the sensed pixel may be further away than the range of the depth sensor, but not the color camera.

So I'm using these equations, where DEPTH is the estimated depth of the point.

x = DEPTH * (u - cx) / fx 
y = DEPTH * (v - cy) / fy 
z = DEPTH

But I'm not sure whether I should be using the unrectifed image or the rectified image? And whether I should be using should be using cx, cy, fx and fy from K or P in CameraInfo.

I'm guessing that if I use the rectified image then the image points will be distorted away from their corresponding depth points... And then if I did cloud.at(u,v) on a point that did have a depth value, then it might not actually find the point.

Thanks

Jamie

Yeah I think its a bit confusing too :S

I'll change the camera xacro link to whatever the offical repo becomes :P

Here's how I'm parametrising the xacro for the camera now if you want to see: https://github.com/jdddog/zoidstein_r...

Yeah that's a great idea, it would make it a lot easier if you could just depend on a package and then link to the xacro of the sensor you were using.

Awesome! Thanks for answering so fast :)

Hi

Has anyone made a URDF for the Creative Senz3D or know where to find one? I'm looking for measurements that describe the relationship between different frames of the Creative Senz3D. Its also known as the Softkinetic DS325V2_M.

Here's an example of a URDF file describing the Kinect's frames of reference:

    <joint name="camera_joint" type="fixed">
        <origin xyz="" rpy="" />
        <parent link="" />
        <child link="camera_link" />
    </joint>

    <joint name="camera_depth_joint" type="fixed">
        <origin xyz="0 0.018 0" rpy="0 0 0" />
        <parent link="camera_link" />
        <child link="camera_depth_frame" />
    </joint>

    <joint name="camera_depth_optical_joint" type="fixed">
        <origin xyz="0 0 0" rpy="${-M_PI/2} 0 ${-M_PI/2}" />
        <parent link="camera_depth_frame" />
        <child link="camera_depth_optical_frame" />
    </joint>

    <joint name="camera_rgb_joint" type="fixed">
        <origin xyz="0 -0.005 0" rpy="0 0 0" />
        <parent link="camera_link" />
        <child link="camera_rgb_frame" />
    </joint>

    <joint name="camera_rgb_optical_joint" type="fixed">
        <origin xyz="0 0 0" rpy="${-M_PI/2} 0 ${-M_PI/2}" />
        <parent link="camera_rgb_frame" />
        <child link="camera_rgb_optical_frame" />
    </joint>

Thanks!

Jamie

Hi

I'm having trouble installting ros indigo on Ubuntu 14.04. I've followed all of the instructions on this page, but when I run sudo apt-get install ros-indigo-desktop-full I get the following errors:

sudo apt-get install ros-indigo-desktop-full
Reading package lists... Done
Building dependency tree       
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 ros-indigo-desktop-full : Depends: ros-indigo-desktop but it is not going to be installed
                           Depends: ros-indigo-simulators but it is not going to be installed
E: Unable to correct problems, you have held broken packages.

I've narrowed the problem down, it won't install because it can't find the package ros-indigo-urdfdom-headers:

user@computer:/etc/apt$ sudo apt-get install ros-indigo-urdfdom-headers
Reading package lists... Done
Building dependency tree       
Reading state information... Done
Package ros-indigo-urdfdom-headers is not available, but is referred to by another package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source

E: Package 'ros-indigo-urdfdom-headers' has no installation candidate

Is there a solution to this?

Thanks

Jamie

You're welcome Ken!

No worries, did you clone the entire directory? e.g. git clone https://github.com/jdddog/people.git And make the stack (not just one package): rosmake people It also needs to be in your rosbuild workspace because I haven't had time to port it to catkin.

Hi Grega

I just ported the leg_detector2 package to work with Hydro and the Kinect:

https://github.com/jdddog/people

Thanks

Jamie

Hi

What options do I have in ROS for controlling a Nao robots body parts using inverse kinematics (e.g. its arms)?

Or would I need to write a wrapper so that ROS can use Naoqi's inverse kinematics solver?

Kind Regards

Jamie