Ask Your Question

Alper Aydemir's profile - activity

2017-08-10 08:19:15 -0500 received badge  Good Question (source)
2016-03-17 05:19:19 -0500 received badge  Good Answer (source)
2014-10-21 04:58:05 -0500 received badge  Taxonomist
2013-08-27 15:19:19 -0500 received badge  Famous Question (source)
2013-07-28 22:53:52 -0500 received badge  Notable Question (source)
2013-06-24 18:04:23 -0500 received badge  Popular Question (source)
2013-06-21 06:45:20 -0500 received badge  Nice Question (source)
2013-06-19 18:03:04 -0500 asked a question Write a tfMessage to bag file

Hi,

I'm trying to write a tfMessage to bag file.

I can very well write a StampedTransform using transformStampedTFToMsg, which helps generating a geometry_msgs::StampedTransform. However this results in an error in a package I'm using:

Client [/mono_odometer] wants topic /tf to have datatype/md5sum [tf/tfMessage/94810edda583a504dfda3829e70d7eec], but our version has [geometry_msgs/TransformStamped/b5764a33bfeb3588febc2682852579b0]. Dropping connection.

And this

tf::tfMessage msg; bag.write("tf", ros::Time(data->timestamp*1e-6), msg);

Does not compile (and there's no way to set a transform for msg).

What's a way of writing tfMessage to bag file while setting a transform and a timestamp?

2013-03-03 21:16:46 -0500 received badge  Famous Question (source)
2013-03-03 21:16:46 -0500 received badge  Popular Question (source)
2013-03-03 21:16:46 -0500 received badge  Notable Question (source)
2012-10-05 21:50:05 -0500 received badge  Famous Question (source)
2012-08-16 05:09:02 -0500 received badge  Famous Question (source)
2012-08-16 05:09:02 -0500 received badge  Popular Question (source)
2012-08-16 05:09:02 -0500 received badge  Notable Question (source)
2012-08-09 12:43:51 -0500 received badge  Good Answer (source)
2012-06-21 05:47:16 -0500 received badge  Nice Answer (source)
2012-03-04 02:38:11 -0500 received badge  Nice Question (source)
2012-01-23 23:05:45 -0500 answered a question Depth map compression

I'm not doing this via ROS, think of it as a separate library. The PNG compression does much better than raw of course but it's not enough for the requirements that I posted. One thing that it misses is the temporal redundancy between consecutive frames.

This topic seems to be a new research question as there are recent papers on it comparing to JPEG 2000 . It seems the only option is to implement those and see.

2012-01-23 02:28:11 -0500 asked a question Depth map compression

Hi,

I'm trying to compress a depth map stream from the Kinect sensor. The requirements is that at 10fps, the uplink should be approx. 1.5 Mbit/s which is most ordinary people have at home (granted, depending on the location). I've tried plain bz2'ing the depth map (~50kb each) and recording the depth map as a smooth gradient in the RGB space and doing plain video compression on it. With ffmpeg lossless compression it's still prohibitively big, but it seemed to get there with lossy compression. Of course it depends on how lossy.

Is there anyone who looked into this? Note that I'm not handling point clouds or tracking the camera position. Anyone experimented with lossy compression rates?

Are there other ways of doing this? The choice of language is C++.

2011-12-14 05:48:30 -0500 received badge  Notable Question (source)
2011-07-28 11:30:13 -0500 received badge  Popular Question (source)
2011-07-11 01:22:20 -0500 answered a question modular_cloud_matcher compilation error

This is now fixed by Stephane Magnenat, git pull to get the changes in libpointmatcher library.

2011-07-11 00:58:31 -0500 asked a question modular_cloud_matcher compilation error

Hi

Trying to get extrinsics between a kinect camera and a robot, I've compiled libpointmatcher and libnao and finally compiling Automatic Calibration of Extrinsic Parameters from here

However modular_cloud_matcher is complaining about a missing member in ICPSequence class in the Core.cpp of pointmatcher library.


/home/cogx/code/ros/ros mapping/modular_cloud_matcher/src/modular_cloud_matcher_bag.cpp:351:27: error: ‘struct MetricSpaceAligner<float>::ICPSequence’ has no member named ‘hasKeyFrame’


I guess this is due to a change in the API.

How should one proceed? Thanks.

2011-04-22 08:51:16 -0500 answered a question turtlebot kinect mounting

Thanks, that is what we do as well and I was wondering if there was some special trick. http://www.csc.kth.se/~aydemir/kinect_strip/kinect.html

2011-04-21 06:36:36 -0500 marked best answer turtlebot kinect mounting

If you pull off the little vent covers on the bottom (they are just a stick on, flexible plastic), you'll expose a total of 4 screws. These screws do require a security Torx driver to remove, but once out, you can thread a standoff into them.

I believe this is what the turtlebot is using. I am using these mount holes on my robot Maxwell.

2011-04-20 01:27:27 -0500 received badge  Student (source)
2011-04-20 01:13:54 -0500 asked a question turtlebot kinect mounting

Hi!

I was wondering how the kinect is mounted on turtlebot as I'm trying to mount one on our Pioneer robot sporting a pan tilt unit. So I thought I could ask a quick question about that.

I can see that perhaps you drilled some holes on Kinect and used just the right screw to squeeze it in (if not how to fasten from the other part?) Or used the existing holes under air vents for the bottom part?

If you have a parts list and some quick instructions for mounting that would be most helpful.

2011-03-16 08:32:03 -0500 received badge  Nice Answer (source)
2011-02-23 19:20:50 -0500 received badge  Supporter (source)
2011-02-18 00:57:28 -0500 received badge  Teacher (source)
2011-02-17 20:09:10 -0500 edited answer Which robot can or should I get to run ROS?

If you want a relatively cheap but still powerful and flexible setup then also you might want to look at Roboard. It is basically a mini-ITX board customized towards robotics needs. We are holding a robotics course here at KTH with students building a robot and writing software for a contest from scratch, they start only with a toolbox, aluminium sheets and wheels + RoBoard + a webcam + Serializer 3.0 + couple of range finding sensors (IR + Sonar mostly).

The end goal is to have a contest where students compete in a simulation of a rescue mission. The robot is in a maze, maps and "rescues" golf balls (victims) in the priority order indicated by a near by barcode from the maze all the while avoiding bombs.

Roboard also has a ROS repository and an OK C++ library. Here's the ROS blog post about Roboard http://www.ros.org/news/2010/10/ros-repository-for-roboard-and-humanoids-now-available.html

here's the course wiki with instructions: http://cogvis.nada.kth.se/roboticscourse/wiki/ Here's a student group video with a robot that they've built from scratch. http://www.youtube.com/watch?v=ZQqyBHkNOvA

Hope this helps.

This year we want to switch to ROS, I guess at some point I should send an email to ROS people when it matures..